task stringclasses 60 values | visual_input_component stringclasses 21 values | source stringclasses 64 values | options stringlengths 12 4.46k | question stringlengths 17 1.33k | context stringlengths 24 21.8k | input_image_path listlengths 1 62 | output stringclasses 10 values |
|---|---|---|---|---|---|---|---|
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_168_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_168_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_168_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_168_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_168_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_169_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_169_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_169_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_169_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_169_4.jpg"
] | C |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_170_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_170_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_170_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_170_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_170_4.jpg"
] | C |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_171_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_171_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_171_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_171_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_171_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_172_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_172_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_172_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_172_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_172_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_173_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_173_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_173_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_173_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_173_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_174_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_174_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_174_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_174_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_174_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_175_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_175_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_175_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_175_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_175_4.jpg"
] | C |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_176_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_176_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_176_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_176_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_176_4.jpg"
] | B |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_177_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_177_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_177_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_177_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_177_4.jpg"
] | C |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_178_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_178_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_178_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_178_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_178_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_179_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_179_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_179_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_179_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_179_4.jpg"
] | C |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_180_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_180_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_180_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_180_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_180_4.jpg"
] | B |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_181_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_181_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_181_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_181_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_181_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_182_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_182_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_182_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_182_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_182_4.jpg"
] | C |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_183_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_183_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_183_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_183_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_183_4.jpg"
] | B |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_184_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_184_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_184_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_184_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_184_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_185_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_185_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_185_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_185_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_185_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_186_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_186_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_186_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_186_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_186_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_187_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_187_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_187_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_187_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_187_4.jpg"
] | D |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_188_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_188_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_188_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_188_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_188_4.jpg"
] | C |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_189_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_189_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_189_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_189_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_189_4.jpg"
] | C |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_190_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_190_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_190_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_190_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_190_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_191_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_191_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_191_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_191_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_191_4.jpg"
] | B |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_192_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_192_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_192_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_192_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_192_4.jpg"
] | B |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_193_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_193_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_193_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_193_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_193_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_194_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_194_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_194_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_194_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_194_4.jpg"
] | B |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_195_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_195_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_195_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_195_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_195_4.jpg"
] | B |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_196_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_196_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_196_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_196_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_196_4.jpg"
] | A |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_197_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_197_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_197_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_197_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_197_4.jpg"
] | D |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_198_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_198_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_198_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_198_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_198_4.jpg"
] | B |
vehicle_retrieval | ['natural_image'] | veri_retrieval | A: The second image
B: The third image
C: The fourth image
D: The fifth image | Please retrieve the most similar vehicle to the query in the candidate.The query image is the first image. | Select from the following choices.
A: The second image
B: The third image
C: The fourth image
D: The fifth image | [
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_199_0.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_199_1.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_199_2.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_199_3.jpg",
"./High-level-obj-semantic/vehicle_retrieval/vehicle_retrieval_199_4.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Scrape". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_0_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_0_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_1_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_1_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_2_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_2_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Lift Something". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_3_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_3_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_4_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_4_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_5_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_5_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Scrape". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_6_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_6_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_7_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_7_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_8_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_8_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_9_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_9_1.jpg"
] | A |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_10_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_10_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Poke". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_11_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_11_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_12_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_12_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_13_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_13_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Scrape". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_14_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_14_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_15_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_15_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_16_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_16_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_17_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_17_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Scoop". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_18_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_18_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_19_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_19_1.jpg"
] | A |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_20_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_20_1.jpg"
] | A |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_21_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_21_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pull out a nail". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_22_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_22_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_23_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_23_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Lift Something". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_24_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_24_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_25_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_25_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_26_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_26_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_27_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_27_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Lift Something". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_28_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_28_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_29_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_29_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_30_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_30_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Scrape". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_31_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_31_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Poke". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_32_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_32_1.jpg"
] | A |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_33_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_33_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_34_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_34_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_35_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_35_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_36_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_36_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Lift Something". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_37_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_37_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_38_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_38_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_39_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_39_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_40_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_40_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_41_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_41_1.jpg"
] | A |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_42_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_42_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_43_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_43_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_44_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_44_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pull out a nail". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_45_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_45_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_46_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_46_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Lift Something". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_47_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_47_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_48_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_48_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pull out a nail". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_49_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_49_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_50_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_50_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Poke". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_51_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_51_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pull out a nail". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_52_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_52_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pull out a nail". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_53_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_53_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_54_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_54_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Scoop". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_55_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_55_1.jpg"
] | A |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_56_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_56_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pull out a nail". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_57_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_57_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_58_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_58_1.jpg"
] | B |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Brush/Dust". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_59_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_59_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_60_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_60_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Poke". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_61_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_61_1.jpg"
] | A |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_62_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_62_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Scrape". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_63_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_63_1.jpg"
] | D |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mash/Pound". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_64_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_64_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pull out a nail". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_65_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_65_1.jpg"
] | C |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Pour". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_66_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_66_1.jpg"
] | A |
functional_correspondence_blink | 2 natural images | blink | A: Point A
B: Point B
C: Point C
D: Point D | Which point is corresponding to the reference point? | Humans can find corresponding points for the same action between different objects. For instance, if a person uses a pot versus a hammer to "Mash Pound", then the handle of the pot will be the corresponding point to the handle of the hammer because they serve the same function for the action -- to hold; and the bottom of the pot will be the corresponding point to the face of the hammer because they both mash the other object.
Given the following two images, a reference point is annotated on the first image, labeled with REF. You are given multiple red-circled points on the right image, choices of "A, B, C, D" are drawn beside each circle. Select from the choices on the second image and find the corresponding point for the reference point, if we use both items for the action: "Mix". Which point is corresponding to the reference point?
Select from the following choices.
A: Point A
B: Point B
C: Point C
D: Point D
| [
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_67_0.jpg",
"./High-level-obj-semantic/functional_correspondence_blink/functional_correspondence_blink_67_1.jpg"
] | A |