# CLIPSeg[[clipseg]]

## 개요[[overview]]

CLIPSeg 모델은 Timo Lüddecke와 Alexander Ecker가 [Image Segmentation Using Text and Image Prompts](https://huggingface.co/papers/2112.10003) 논문에서 제안했습니다. CLIPSeg는 가중치가 고정된 CLIP 모델에 최소한의 디코더를 결합하여 제로샷 및 원샷 이미지 분할을 수행합니다.

논문 초록은 다음과 같습니다.

*이미지 분할은 일반적으로 사전에 정의된 객체 클래스 집합에 대해 모델을 훈련시키는 방식으로 접근합니다. 하지만 새로운 클래스를 추가하거나 보다 복잡한 질의를 처리하려면, 해당 내용을 포함한 데이터 세트로 모델을 다시 훈련해야 하므로 비용이 많이 듭니다. 이에 본 논문에서는 테스트 시점에 텍스트나 이미지로 구성된 임의의 프롬프트만으로 이미지 분할을 수행할 수 있는 시스템을 제안합니다. 이 접근 방식을 통해 서로 다른 과제를 갖는 세 가지 주요 이미지 분할 태스크—지시 표현 분할(referring expression segmentation), 제로샷 분할(zero-shot segmentation), 원샷 분할(one-shot segmentation)—을 단일 통합 모델로 처리할 수 있습니다. 이를 위해 우리는 CLIP 모델을 백본으로 삼고, 고해상도 예측을 가능하게 하는 트랜스포머 기반 디코더를 추가해 이를 확장했습니다. 확장된 PhraseCut 데이터 세트를 활용해 훈련한 본 시스템은 자유 형식의 텍스트 프롬프트나 특정 목적을 표현하는 이미지를 입력으로 받아, 입력 이미지에 대한 이진 분할 맵을 생성합니다. 특히 이미지 기반 프롬프트의 다양한 구성 방식과 그 효과를 자세히 분석하였습니다. 이 새로운 하이브리드 입력 방식은 앞서 언급한 세 가지 태스크뿐만 아니라, 텍스트 또는 이미지로 질의할 수 있는 모든 이진 분할 문제에 유연하게 대응할 수 있습니다. 마지막으로, 본 시스템이 어포던스(affordance)나 객체 속성과 같은 일반화된 질의에도 높은 적응력을 보임을 확인하였습니다.*

 

 CLIPSeg 개요. 원본 논문에서 발췌. 

이 모델은 [nielsr](https://huggingface.co/nielsr)님이 기여했습니다.
원본 코드는 [여기](https://github.com/timojl/clipseg)에서 찾을 수 있습니다.

## 사용 팁[[usage-tips]]

- [CLIPSegForImageSegmentation](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegForImageSegmentation)은 [CLIPSegModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegModel)과 동일한, [CLIPSegModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegModel) 위에 디코더를 추가한 모델입니다.
- [CLIPSegForImageSegmentation](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegForImageSegmentation)은 테스트 시점에 임의의 프롬프트를 기반으로 이미지 분할을 생성합니다. 이때 프롬프트는 텍스트(`input_ids`), 이미지(`conditional_pixel_values`), 사용자 정의 조건부 임베딩(`conditional_embeddings`)을 사용할 수 있습니다.

## 리소스[[resources]]

CLIPSeg를 시작하는 데 도움이 될 Hugging Face 공식 자료와 커뮤니티(🌎 아이콘으로 표시)의 유용한 리소스 목록을 아래에 정리했습니다. 혹시 목록에 없는 새로운 자료나 튜토리얼을 공유하고 싶으시다면, 언제든지 Pull Request를 통해 제안해 주세요. 저희가 검토 후 소중히 반영하겠습니다! 기존 자료와 중복되지 않는 새로운 내용이라면 더욱 좋습니다.

- [zero-shot image segmentation with CLIPSeg](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/CLIPSeg/Zero_shot_image_segmentation_with_CLIPSeg.ipynb)을 시연하는 노트북.

## CLIPSegConfig[[transformers.CLIPSegConfig]][[transformers.CLIPSegConfig]]

#### transformers.CLIPSegConfig[[transformers.CLIPSegConfig]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/configuration_clipseg.py#L124)

This is the configuration class to store the configuration of a CLIPSegModel. It is used to instantiate a Clipseg
model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
defaults will yield a similar configuration to that of the [CIDAS/clipseg-rd64](https://huggingface.co/CIDAS/clipseg-rd64)

Configuration objects inherit from [PreTrainedConfig](/docs/transformers/v5.6.2/ko/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the
documentation from [PreTrainedConfig](/docs/transformers/v5.6.2/ko/main_classes/configuration#transformers.PreTrainedConfig) for more information.

Example:

```python
>>> from transformers import CLIPSegConfig, CLIPSegModel

>>> # Initializing a CLIPSegConfig with CIDAS/clipseg-rd64 style configuration
>>> configuration = CLIPSegConfig()

>>> # Initializing a CLIPSegModel (with random weights) from the CIDAS/clipseg-rd64 style configuration
>>> model = CLIPSegModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

>>> # We can also initialize a CLIPSegConfig from a CLIPSegTextConfig and a CLIPSegVisionConfig

>>> # Initializing a CLIPSegText and CLIPSegVision configuration
>>> config_text = CLIPSegTextConfig()
>>> config_vision = CLIPSegVisionConfig()

>>> config = CLIPSegConfig(text_config=config_text, vision_config=config_vision)
```

**Parameters:**

text_config (`Union[dict, ~models.clipseg.configuration_clipseg.CLIPSegTextConfig]`, *optional*) : The config object or dictionary of the text backbone.

vision_config (`Union[dict, ~models.clipseg.configuration_clipseg.CLIPSegVisionConfig]`, *optional*) : The config object or dictionary of the vision backbone.

projection_dim (`int`, *optional*, defaults to `512`) : Dimensionality of text and vision projection layers.

logit_scale_init_value (`Union[float, int]`, *optional*, defaults to `2.6592`) : The initial value of the *logit_scale* parameter.

initializer_factor (`float`, *optional*, defaults to `1.0`) : A factor for initializing all weight matrices (should be kept to 1, used internally for initialization testing).

extract_layers (`list[int]`, *optional*, defaults to `[3, 6, 9]`) : Layers to extract when forwarding the query image through the frozen visual backbone of CLIP.

reduce_dim (`int`, *optional*, defaults to 64) : Dimensionality to reduce the CLIP vision embedding.

decoder_num_attention_heads (`int`, *optional*, defaults to `4`) : Number of attention heads for each attention layer in the Transformer decoder.

decoder_attention_dropout (`Union[float, int]`, *optional*, defaults to `0.0`) : The dropout ratio for the attention probabilities.

decoder_hidden_act (`str`, *optional*, defaults to `quick_gelu`) : The non-linear activation function (function or string) in the decoder. For example, `"gelu"`, `"relu"`, `"silu"`, etc.

decoder_intermediate_size (`int`, *optional*, defaults to `2048`) : Dimension of the MLP representations.

conditional_layer (`int`, *optional*, defaults to 0) : The layer to use of the Transformer encoder whose activations will be combined with the condition embeddings using FiLM (Feature-wise Linear Modulation). If 0, the last layer is used.

use_complex_transposed_convolution (`bool`, *optional*, defaults to `False`) : Whether to use a more complex transposed convolution in the decoder, enabling more fine-grained segmentation..

## CLIPSegTextConfig[[transformers.CLIPSegTextConfig]][[transformers.CLIPSegTextConfig]]

#### transformers.CLIPSegTextConfig[[transformers.CLIPSegTextConfig]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/configuration_clipseg.py#L31)

This is the configuration class to store the configuration of a CLIPSegModel. It is used to instantiate a Clipseg
model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
defaults will yield a similar configuration to that of the [CIDAS/clipseg-rd64](https://huggingface.co/CIDAS/clipseg-rd64)

Configuration objects inherit from [PreTrainedConfig](/docs/transformers/v5.6.2/ko/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the
documentation from [PreTrainedConfig](/docs/transformers/v5.6.2/ko/main_classes/configuration#transformers.PreTrainedConfig) for more information.

Example:

```python
>>> from transformers import CLIPSegTextConfig, CLIPSegTextModel

>>> # Initializing a CLIPSegTextConfig with CIDAS/clipseg-rd64 style configuration
>>> configuration = CLIPSegTextConfig()

>>> # Initializing a CLIPSegTextModel (with random weights) from the CIDAS/clipseg-rd64 style configuration
>>> model = CLIPSegTextModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config
```

**Parameters:**

vocab_size (`int`, *optional*, defaults to `49408`) : Vocabulary size of the model. Defines the number of different tokens that can be represented by the `input_ids`.

hidden_size (`int`, *optional*, defaults to `512`) : Dimension of the hidden representations.

intermediate_size (`int`, *optional*, defaults to `2048`) : Dimension of the MLP representations.

num_hidden_layers (`int`, *optional*, defaults to `12`) : Number of hidden layers in the Transformer decoder.

num_attention_heads (`int`, *optional*, defaults to `8`) : Number of attention heads for each attention layer in the Transformer decoder.

max_position_embeddings (`int`, *optional*, defaults to `77`) : The maximum sequence length that this model might ever be used with.

hidden_act (`str`, *optional*, defaults to `quick_gelu`) : The non-linear activation function (function or string) in the decoder. For example, `"gelu"`, `"relu"`, `"silu"`, etc.

layer_norm_eps (`float`, *optional*, defaults to `1e-05`) : The epsilon used by the layer normalization layers.

attention_dropout (`Union[int, float]`, *optional*, defaults to `0.0`) : The dropout ratio for the attention probabilities.

initializer_range (`float`, *optional*, defaults to `0.02`) : The standard deviation of the truncated_normal_initializer for initializing all weight matrices.

initializer_factor (`float`, *optional*, defaults to `1.0`) : A factor for initializing all weight matrices (should be kept to 1, used internally for initialization testing).

pad_token_id (`int`, *optional*, defaults to `1`) : Token id used for padding in the vocabulary.

bos_token_id (`int`, *optional*, defaults to `49406`) : Token id used for beginning-of-stream in the vocabulary.

eos_token_id (`Union[int, list[int]]`, *optional*, defaults to `49407`) : Token id used for end-of-stream in the vocabulary.

## CLIPSegVisionConfig[[transformers.CLIPSegVisionConfig]][[transformers.CLIPSegVisionConfig]]

#### transformers.CLIPSegVisionConfig[[transformers.CLIPSegVisionConfig]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/configuration_clipseg.py#L80)

This is the configuration class to store the configuration of a CLIPSegModel. It is used to instantiate a Clipseg
model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
defaults will yield a similar configuration to that of the [CIDAS/clipseg-rd64](https://huggingface.co/CIDAS/clipseg-rd64)

Configuration objects inherit from [PreTrainedConfig](/docs/transformers/v5.6.2/ko/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the
documentation from [PreTrainedConfig](/docs/transformers/v5.6.2/ko/main_classes/configuration#transformers.PreTrainedConfig) for more information.

Example:

```python
>>> from transformers import CLIPSegVisionConfig, CLIPSegVisionModel

>>> # Initializing a CLIPSegVisionConfig with CIDAS/clipseg-rd64 style configuration
>>> configuration = CLIPSegVisionConfig()

>>> # Initializing a CLIPSegVisionModel (with random weights) from the CIDAS/clipseg-rd64 style configuration
>>> model = CLIPSegVisionModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config
```

**Parameters:**

hidden_size (`int`, *optional*, defaults to `768`) : Dimension of the hidden representations.

intermediate_size (`int`, *optional*, defaults to `3072`) : Dimension of the MLP representations.

num_hidden_layers (`int`, *optional*, defaults to `12`) : Number of hidden layers in the Transformer decoder.

num_attention_heads (`int`, *optional*, defaults to `12`) : Number of attention heads for each attention layer in the Transformer decoder.

num_channels (`int`, *optional*, defaults to `3`) : The number of input channels.

image_size (`Union[int, list[int], tuple[int, int]]`, *optional*, defaults to `224`) : The size (resolution) of each image.

patch_size (`Union[int, list[int], tuple[int, int]]`, *optional*, defaults to `32`) : The size (resolution) of each patch.

hidden_act (`str`, *optional*, defaults to `quick_gelu`) : The non-linear activation function (function or string) in the decoder. For example, `"gelu"`, `"relu"`, `"silu"`, etc.

layer_norm_eps (`float`, *optional*, defaults to `1e-05`) : The epsilon used by the layer normalization layers.

attention_dropout (`Union[int, float]`, *optional*, defaults to `0.0`) : The dropout ratio for the attention probabilities.

initializer_range (`float`, *optional*, defaults to `0.02`) : The standard deviation of the truncated_normal_initializer for initializing all weight matrices.

initializer_factor (`float`, *optional*, defaults to `1.0`) : A factor for initializing all weight matrices (should be kept to 1, used internally for initialization testing).

## CLIPSegProcessor[[transformers.CLIPSegProcessor]][[transformers.CLIPSegProcessor]]

#### transformers.CLIPSegProcessor[[transformers.CLIPSegProcessor]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/processing_clipseg.py#L24)

Constructs a CLIPSegProcessor which wraps a image processor and a tokenizer into a single processor.

[CLIPSegProcessor](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegProcessor) offers all the functionalities of [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor) and [CLIPTokenizer](/docs/transformers/v5.6.2/ko/model_doc/clip#transformers.CLIPTokenizer). See the
[~ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor) and [~CLIPTokenizer](/docs/transformers/v5.6.2/ko/model_doc/clip#transformers.CLIPTokenizer) for more information.

**Parameters:**

image_processor (`ViTImageProcessor`) : The image processor is a required input.

tokenizer (`CLIPTokenizer`) : The tokenizer is a required input.

## CLIPSegModel[[transformers.CLIPSegModel]][[transformers.CLIPSegModel]]

#### transformers.CLIPSegModel[[transformers.CLIPSegModel]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L789)

The bare Clipseg Model outputting raw hidden-states without any specific head on top.

This model inherits from [PreTrainedModel](/docs/transformers/v5.6.2/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)

This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.

forwardtransformers.CLIPSegModel.forwardhttps://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L883[{"name": "input_ids", "val": ": torch.LongTensor | None = None"}, {"name": "pixel_values", "val": ": torch.FloatTensor | None = None"}, {"name": "attention_mask", "val": ": torch.Tensor | None = None"}, {"name": "position_ids", "val": ": torch.LongTensor | None = None"}, {"name": "return_loss", "val": ": bool | None = None"}, {"name": "interpolate_pos_encoding", "val": ": bool = True"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}]- **input_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

  Indices can be obtained using [AutoTokenizer](/docs/transformers/v5.6.2/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.6.2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and
  [PreTrainedTokenizer.__call__()](/docs/transformers/v5.6.2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details.

  [What are input IDs?](../glossary#input-ids)
- **pixel_values** (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`, *optional*) --
  The tensors corresponding to the input images. Pixel values can be obtained using
  [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor). See `ViTImageProcessor.__call__()` for details ([CLIPSegProcessor](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegProcessor) uses
  [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor) for processing images).
- **attention_mask** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:

  - 1 for tokens that are **not masked**,
  - 0 for tokens that are **masked**.

  [What are attention masks?](../glossary#attention-mask)
- **position_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`.

  [What are position IDs?](../glossary#position-ids)
- **return_loss** (`bool`, *optional*) --
  Whether or not to return the contrastive loss.
- **interpolate_pos_encoding** (`bool`, *optional*, defaults to `True`) --
  Whether to interpolate the pre-trained position encodings.0`CLIPSegOutput` or `tuple(torch.FloatTensor)`A `CLIPSegOutput` or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.
The [CLIPSegModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegModel) forward method, overrides the `__call__` special method.

Although the recipe for forward pass needs to be defined within this function, one should call the `Module`
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.

- **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `return_loss` is `True`) -- Contrastive loss for image-text similarity.
- **logits_per_image** (`torch.FloatTensor` of shape `(image_batch_size, text_batch_size)`) -- The scaled dot product scores between `image_embeds` and `text_embeds`. This represents the image-text
  similarity scores.
- **logits_per_text** (`torch.FloatTensor` of shape `(text_batch_size, image_batch_size)`) -- The scaled dot product scores between `text_embeds` and `image_embeds`. This represents the text-image
  similarity scores.
- **text_embeds** (`torch.FloatTensor` of shape `(batch_size, output_dim`) -- The text embeddings obtained by applying the projection layer to the pooled output of [CLIPSegTextModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegTextModel).
- **image_embeds** (`torch.FloatTensor` of shape `(batch_size, output_dim`) -- The image embeddings obtained by applying the projection layer to the pooled output of [CLIPSegVisionModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegVisionModel).
- **text_model_output** (`~modeling_outputs.BaseModelOutputWithPooling`, defaults to `None`) -- The output of the [CLIPSegTextModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegTextModel).
- **vision_model_output** (`~modeling_outputs.BaseModelOutputWithPooling`, defaults to `None`) -- The output of the [CLIPSegVisionModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegVisionModel).

Examples:

```python
>>> import torch
>>> from transformers import AutoProcessor, CLIPSegModel
>>> from transformers.image_utils import load_image

>>> processor = AutoProcessor.from_pretrained("CIDAS/clipseg-rd64-refined")
>>> model = CLIPSegModel.from_pretrained("CIDAS/clipseg-rd64-refined")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = load_image(url)

>>> inputs = processor(
...     text=["a photo of a cat", "a photo of a dog"], images=image, return_tensors="pt", padding=True
... )

>>> with torch.inference_mode():
...     outputs = model(**inputs)
>>> logits_per_image = outputs.logits_per_image  # this is the image-text similarity score
>>> probs = logits_per_image.softmax(dim=1)  # we can take the softmax to get the label probabilities
```

**Parameters:**

config ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v5.6.2/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

**Returns:**

``CLIPSegOutput` or `tuple(torch.FloatTensor)``

A `CLIPSegOutput` or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.
#### get_text_features[[transformers.CLIPSegModel.get_text_features]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L810)

- **last_hidden_state** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`) -- Sequence of hidden-states at the output of the last layer of the model.
- **pooler_output** (`torch.FloatTensor` of shape `(batch_size, hidden_size)`) -- Last layer hidden-state of the first token of the sequence (classification token) after further processing
  through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns
  the classification token after processing through a linear layer and a tanh activation function. The linear
  layer weights are trained from the next sentence prediction (classification) objective during pretraining.
- **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
  one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.

  Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
- **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
  sequence_length)`.

  Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
  heads.

Examples:

```python
>>> import torch
>>> from transformers import AutoTokenizer, CLIPSegModel

>>> tokenizer = AutoTokenizer.from_pretrained("CIDAS/clipseg-rd64-refined")
>>> model = CLIPSegModel.from_pretrained("CIDAS/clipseg-rd64-refined")

>>> inputs = tokenizer(["a photo of a cat", "a photo of a dog"], padding=True, return_tensors="pt")
>>> with torch.inference_mode():
...     text_features = model.get_text_features(**inputs)
```

**Parameters:**

input_ids (`torch.Tensor` of shape `(batch_size, sequence_length)`) : Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.  Indices can be obtained using [AutoTokenizer](/docs/transformers/v5.6.2/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.6.2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.6.2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details.  [What are input IDs?](../glossary#input-ids)

attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) : Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:  - 1 for tokens that are **not masked**, - 0 for tokens that are **masked**.  [What are attention masks?](../glossary#attention-mask)

position_ids (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) : Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`.  [What are position IDs?](../glossary#position-ids)

**Returns:**

`[BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)``

A [BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.
#### get_image_features[[transformers.CLIPSegModel.get_image_features]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L845)

- **last_hidden_state** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`) -- Sequence of hidden-states at the output of the last layer of the model.
- **pooler_output** (`torch.FloatTensor` of shape `(batch_size, hidden_size)`) -- Last layer hidden-state of the first token of the sequence (classification token) after further processing
  through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns
  the classification token after processing through a linear layer and a tanh activation function. The linear
  layer weights are trained from the next sentence prediction (classification) objective during pretraining.
- **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
  one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.

  Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
- **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
  sequence_length)`.

  Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
  heads.

Examples:

```python
>>> import torch
>>> from transformers import AutoProcessor, CLIPSegModel
>>> from transformers.image_utils import load_image

>>> processor = AutoProcessor.from_pretrained("CIDAS/clipseg-rd64-refined")
>>> model = CLIPSegModel.from_pretrained("CIDAS/clipseg-rd64-refined")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = load_image(url)

>>> inputs = processor(images=image, return_tensors="pt")

>>> with torch.inference_mode():
...     image_features = model.get_image_features(**inputs)
```

**Parameters:**

pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`) : The tensors corresponding to the input images. Pixel values can be obtained using [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor). See `ViTImageProcessor.__call__()` for details ([CLIPSegProcessor](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegProcessor) uses [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor) for processing images).

interpolate_pos_encoding (`bool`, *optional*, defaults to `True`) : Whether to interpolate the pre-trained position encodings.

**Returns:**

`[BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)``

A [BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.

## CLIPSegTextModel[[transformers.CLIPSegTextModel]][[transformers.CLIPSegTextModel]]

#### transformers.CLIPSegTextModel[[transformers.CLIPSegTextModel]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L600)

The text model from CLIPSEG without any head or projection on top.

This model inherits from [PreTrainedModel](/docs/transformers/v5.6.2/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)

This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.

forwardtransformers.CLIPSegTextModel.forwardhttps://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L616[{"name": "input_ids", "val": ": torch.Tensor | None = None"}, {"name": "attention_mask", "val": ": torch.Tensor | None = None"}, {"name": "position_ids", "val": ": torch.Tensor | None = None"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}]- **input_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

  Indices can be obtained using [AutoTokenizer](/docs/transformers/v5.6.2/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.6.2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and
  [PreTrainedTokenizer.__call__()](/docs/transformers/v5.6.2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details.

  [What are input IDs?](../glossary#input-ids)
- **attention_mask** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:

  - 1 for tokens that are **not masked**,
  - 0 for tokens that are **masked**.

  [What are attention masks?](../glossary#attention-mask)
- **position_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`.

  [What are position IDs?](../glossary#position-ids)0[BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)`A [BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.
The [CLIPSegTextModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegTextModel) forward method, overrides the `__call__` special method.

Although the recipe for forward pass needs to be defined within this function, one should call the `Module`
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.

- **last_hidden_state** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`) -- Sequence of hidden-states at the output of the last layer of the model.
- **pooler_output** (`torch.FloatTensor` of shape `(batch_size, hidden_size)`) -- Last layer hidden-state of the first token of the sequence (classification token) after further processing
  through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns
  the classification token after processing through a linear layer and a tanh activation function. The linear
  layer weights are trained from the next sentence prediction (classification) objective during pretraining.
- **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
  one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.

  Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
- **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
  sequence_length)`.

  Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
  heads.

Examples:

```python
>>> from transformers import AutoTokenizer, CLIPSegTextModel

>>> tokenizer = AutoTokenizer.from_pretrained("CIDAS/clipseg-rd64-refined")
>>> model = CLIPSegTextModel.from_pretrained("CIDAS/clipseg-rd64-refined")

>>> inputs = tokenizer(["a photo of a cat", "a photo of a dog"], padding=True, return_tensors="pt")

>>> outputs = model(**inputs)
>>> last_hidden_state = outputs.last_hidden_state
>>> pooled_output = outputs.pooler_output  # pooled (EOS token) states
```

**Parameters:**

config ([CLIPSegTextConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegTextConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v5.6.2/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

**Returns:**

`[BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)``

A [BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.

## CLIPSegVisionModel[[transformers.CLIPSegVisionModel]][[transformers.CLIPSegVisionModel]]

#### transformers.CLIPSegVisionModel[[transformers.CLIPSegVisionModel]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L700)

The vision model from CLIPSEG without any head or projection on top.

This model inherits from [PreTrainedModel](/docs/transformers/v5.6.2/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)

This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.

forwardtransformers.CLIPSegVisionModel.forwardhttps://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L716[{"name": "pixel_values", "val": ": torch.FloatTensor | None"}, {"name": "interpolate_pos_encoding", "val": ": bool | None = True"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}]- **pixel_values** (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`, *optional*) --
  The tensors corresponding to the input images. Pixel values can be obtained using
  [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor). See `ViTImageProcessor.__call__()` for details ([CLIPSegProcessor](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegProcessor) uses
  [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor) for processing images).
- **interpolate_pos_encoding** (`bool`, *optional*, defaults to `True`) --
  Whether to interpolate the pre-trained position encodings.0[BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)`A [BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.
The [CLIPSegVisionModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegVisionModel) forward method, overrides the `__call__` special method.

Although the recipe for forward pass needs to be defined within this function, one should call the `Module`
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.

- **last_hidden_state** (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`) -- Sequence of hidden-states at the output of the last layer of the model.
- **pooler_output** (`torch.FloatTensor` of shape `(batch_size, hidden_size)`) -- Last layer hidden-state of the first token of the sequence (classification token) after further processing
  through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns
  the classification token after processing through a linear layer and a tanh activation function. The linear
  layer weights are trained from the next sentence prediction (classification) objective during pretraining.
- **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
  one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.

  Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
- **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
  sequence_length)`.

  Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
  heads.

Examples:

```python
>>> import httpx
>>> from io import BytesIO
>>> from PIL import Image
>>> from transformers import AutoProcessor, CLIPSegVisionModel

>>> processor = AutoProcessor.from_pretrained("CIDAS/clipseg-rd64-refined")
>>> model = CLIPSegVisionModel.from_pretrained("CIDAS/clipseg-rd64-refined")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> with httpx.stream("GET", url) as response:
...     image = Image.open(BytesIO(response.read()))

>>> inputs = processor(images=image, return_tensors="pt")

>>> outputs = model(**inputs)
>>> last_hidden_state = outputs.last_hidden_state
>>> pooled_output = outputs.pooler_output  # pooled CLS states
```

**Parameters:**

config ([CLIPSegVisionConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegVisionConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v5.6.2/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

**Returns:**

`[BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)``

A [BaseModelOutputWithPooling](/docs/transformers/v5.6.2/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.

## CLIPSegForImageSegmentation[[transformers.CLIPSegForImageSegmentation]][[transformers.CLIPSegForImageSegmentation]]

#### transformers.CLIPSegForImageSegmentation[[transformers.CLIPSegForImageSegmentation]]

[Source](https://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L967)

CLIPSeg model with a Transformer-based decoder on top for zero-shot and one-shot image segmentation.

This model inherits from [PreTrainedModel](/docs/transformers/v5.6.2/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)

This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.

forwardtransformers.CLIPSegForImageSegmentation.forwardhttps://github.com/huggingface/transformers/blob/v5.6.2/src/transformers/models/clipseg/modeling_clipseg.py#L1007[{"name": "input_ids", "val": ": torch.FloatTensor | None = None"}, {"name": "pixel_values", "val": ": torch.FloatTensor | None = None"}, {"name": "conditional_pixel_values", "val": ": torch.FloatTensor | None = None"}, {"name": "conditional_embeddings", "val": ": torch.FloatTensor | None = None"}, {"name": "attention_mask", "val": ": torch.Tensor | None = None"}, {"name": "position_ids", "val": ": torch.LongTensor | None = None"}, {"name": "labels", "val": ": torch.LongTensor | None = None"}, {"name": "interpolate_pos_encoding", "val": ": bool = True"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}]- **input_ids** (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

  Indices can be obtained using [AutoTokenizer](/docs/transformers/v5.6.2/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.6.2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and
  [PreTrainedTokenizer.__call__()](/docs/transformers/v5.6.2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details.

  [What are input IDs?](../glossary#input-ids)
- **pixel_values** (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`, *optional*) --
  The tensors corresponding to the input images. Pixel values can be obtained using
  [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor). See `ViTImageProcessor.__call__()` for details ([CLIPSegProcessor](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegProcessor) uses
  [ViTImageProcessor](/docs/transformers/v5.6.2/ko/model_doc/vit#transformers.ViTImageProcessor) for processing images).
- **conditional_pixel_values** (`torch.FloatTensor`, *optional*) --
  The pixel values of the conditional images.
- **conditional_embeddings** (`torch.FloatTensor` of shape `(batch_size, config.projection_dim)`, *optional*) --
  The conditional embeddings for the query images. If provided, the model will use this instead of computing
  the embeddings from the conditional_pixel_values.
- **attention_mask** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:

  - 1 for tokens that are **not masked**,
  - 0 for tokens that are **masked**.

  [What are attention masks?](../glossary#attention-mask)
- **position_ids** (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0, config.n_positions - 1]`.

  [What are position IDs?](../glossary#position-ids)
- **labels** (`torch.LongTensor` of shape `(batch_size,)`, *optional*) --
  Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
  config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
  `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
- **interpolate_pos_encoding** (`bool`, *optional*, defaults to `True`) --
  Whether to interpolate the pre-trained position encodings.0`CLIPSegOutput` or `tuple(torch.FloatTensor)`A `CLIPSegOutput` or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.
The [CLIPSegForImageSegmentation](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegForImageSegmentation) forward method, overrides the `__call__` special method.

Although the recipe for forward pass needs to be defined within this function, one should call the `Module`
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.

- **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `return_loss` is `True`) -- Contrastive loss for image-text similarity.
- **logits_per_image** (`torch.FloatTensor` of shape `(image_batch_size, text_batch_size)`) -- The scaled dot product scores between `image_embeds` and `text_embeds`. This represents the image-text
  similarity scores.
- **logits_per_text** (`torch.FloatTensor` of shape `(text_batch_size, image_batch_size)`) -- The scaled dot product scores between `text_embeds` and `image_embeds`. This represents the text-image
  similarity scores.
- **text_embeds** (`torch.FloatTensor` of shape `(batch_size, output_dim`) -- The text embeddings obtained by applying the projection layer to the pooled output of [CLIPSegTextModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegTextModel).
- **image_embeds** (`torch.FloatTensor` of shape `(batch_size, output_dim`) -- The image embeddings obtained by applying the projection layer to the pooled output of [CLIPSegVisionModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegVisionModel).
- **text_model_output** (`~modeling_outputs.BaseModelOutputWithPooling`, defaults to `None`) -- The output of the [CLIPSegTextModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegTextModel).
- **vision_model_output** (`~modeling_outputs.BaseModelOutputWithPooling`, defaults to `None`) -- The output of the [CLIPSegVisionModel](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegVisionModel).

Examples:

```python
>>> import torch
>>> from transformers import AutoProcessor, CLIPSegForImageSegmentation
>>> from transformers.image_utils import load_image

>>> processor = AutoProcessor.from_pretrained("CIDAS/clipseg-rd64-refined")
>>> model = CLIPSegForImageSegmentation.from_pretrained("CIDAS/clipseg-rd64-refined")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = load_image(url)

>>> texts = ["a cat", "a remote", "a blanket"]
>>> inputs = processor(text=texts, images=[image] * len(texts), padding=True, return_tensors="pt")

>>> with torch.inference_mode():
...     outputs = model(**inputs)

>>> logits = outputs.logits
>>> print(logits.shape)
torch.Size([3, 352, 352])
```

**Parameters:**

config ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v5.6.2/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

**Returns:**

``CLIPSegOutput` or `tuple(torch.FloatTensor)``

A `CLIPSegOutput` or a tuple of
`torch.FloatTensor` (if `return_dict=False` is passed or when `config.return_dict=False`) comprising various
elements depending on the configuration ([CLIPSegConfig](/docs/transformers/v5.6.2/ko/model_doc/clipseg#transformers.CLIPSegConfig)) and inputs.