Update README.md

README.md

@@ -5,14 +5,16 @@ license: apache-2.0
 # MM-Omni3D
 
 ## Dataset Description
-- **Webpage:** [Dataset Homepage](#)
-- **Paper:** [Dataset Repository](#)
-- **Code:** [https://github.com/Lizhuoling/UniMODE](#)
+- **Webpage:** [Webpage link (to be released)](#)
+- **UniMODE Paper:** [Paper link](https://openaccess.thecvf.com/content/CVPR2024/html/Li_UniMODE_Unified_Monocular_3D_Object_Detection_CVPR_2024_paper.html)
+- **MM-UniMODE Paper:** [Paper link (to be released)](#)
+- **Code:** [Code link](https://github.com/Lizhuoling/UniMODE)
 
 ## Dataset Summary
 MM-UniMODE is the largest multi-modal 3D object detection dataset. It consists of six datasets, i.e., SUN-RGBD, 
 ARKitScenes, Objectron, Hypersim, KITTI, and nuScenes. Among them, SUN-RGBD, ARKitScenes, Objectron, and Hypersim are indoor datasets, 
 while KITTI and nuScenes are outdoor datasets. Besides, Hypersim is a synthesized dataset and the other five datasets are collected from 
 real scenes by different sensors. The MM-Omni3D dataset consists of a total of 234,152 data samples and is divided into 6 subsets, 
-each corresponding to a different data scenario as shown in Fig. 1 (e)∼(j). Among these subsets, four are dedicated to indoor object 
-detection, while the remaining two pertain to outdoor scenes. Each subset is further divided into training, validation, and testing sets.
+each corresponding to a different data scenario. Among these subsets, four are dedicated to indoor object 
+detection, while the remaining two pertain to outdoor scenes. Each subset is further divided into training, validation, and testing sets.
+Please download the whole dataset and follow the official Github code repo to understand how to use it.