Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[SuperPoint, PaliGemma] Update docs #31025

Merged
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
42 changes: 41 additions & 1 deletion docs/source/en/model_doc/paligemma.md
Original file line number Diff line number Diff line change
Expand Up @@ -18,11 +18,51 @@ rendered properly in your Markdown viewer.

## Overview

The PaliGemma model was proposed by Google. It is a 3B VLM composed by a Siglip-400m vision encoder and a Gemma-2B decoder linked by a multimodal linear projection. It is not a chat model with images. It cuts an image into a fixed number of VIT tokens and prepends it to an optional prompt. One particularity is that the model uses full block attention on all the image tokens plus the input text tokens. It comes in 3 resolutions, 224x224, 448x448 and 896x896 with 3 base models, with 55 fine-tuned versions for different tasks, and 2 mix models.
The PaliGemma model was proposed in [PaliGemma – Google's Cutting-Edge Open Vision Language Model](https://huggingface.co/blog/paligemma) by Google. It is a 3B vision-language model composed by a [SigLIP](siglip) vision encoder and a [Gemma](gemma) language decoder linked by a multimodal linear projection. It cuts an image into a fixed number of VIT tokens and prepends it to an optional prompt. One particularity is that the model uses full block attention on all the image tokens plus the input text tokens. It comes in 3 resolutions, 224x224, 448x448 and 896x896 with 3 base models, with 55 fine-tuned versions for different tasks, and 2 mix models.

<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/blog/paligemma/paligemma_arch.png"
alt="drawing" width="600"/>

<small> PaliGemma architecture. Taken from the <a href="https://huggingface.co/blog/paligemma">blog post.</a> </small>

This model was contributed by [Molbap](https://huggingface.co/Molbap).

## Usage tips

Inference with PaliGemma can be performed as follows:

```python
from transformers import AutoProcessor, PaliGemmaForConditionalGeneration

model_id = "google/paligemma-3b-mix-224"
model = PaliGemmaForConditionalGeneration.from_pretrained(model_id)
processor = AutoProcessor.from_pretrained(model_id)

prompt = "What is on the flower?"
image_file = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg?download=true"
raw_image = Image.open(requests.get(image_file, stream=True).raw)
inputs = processor(prompt, raw_image, return_tensors="pt")
output = model.generate(**inputs, max_new_tokens=20)

print(processor.decode(output[0], skip_special_tokens=True)[len(prompt):])
```

- PaliGemma is not meant for conversational use, and it works best when fine-tuning to a specific use case. Some downstream tasks on which PaliGemma can be fine-tuned include image captioning, visual question answering (VQA), object detection, referring expression segmentation and document understanding.
- One can use `PaliGemmaProcessor` to prepare images, text and optional labels for the model. When fine-tuning a PaliGemma model, the `suffix` argument can be passed to the processor which creates the `labels` for the model:

```python
prompt = "What is on the flower?"
answer = "a bee"
inputs = processor(text=prompt, images=raw_image, suffix=answer, return_tensors="pt")
```

## Resources

A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with PaliGemma. If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.

- A blog post introducing all the features of PaliGemma can be found [here](https://huggingface.co/blog/paligemma).
- Demo notebooks on how to fine-tune PaliGemma for VQA with the Trainer API along with inference can be found [here](https://github.com/huggingface/notebooks/tree/main/examples/paligemma).
- Demo notebooks on how to fine-tune PaliGemma on a custom dataset (receipt image -> JSON) along with inference can be found [here](https://github.com/NielsRogge/Transformers-Tutorials/tree/master/PaliGemma). 🌎

## PaliGemmaConfig

Expand Down
21 changes: 16 additions & 5 deletions docs/source/en/model_doc/superpoint.md
Original file line number Diff line number Diff line change
Expand Up @@ -38,12 +38,17 @@ to repeatedly detect a much richer set of interest points than the initial pre-a
traditional corner detector. The final system gives rise to state-of-the-art homography estimation results on HPatches
when compared to LIFT, SIFT and ORB.*

## How to use
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/superpoint_architecture.png"
alt="drawing" width="500"/>

<small> SuperPoint overview. Taken from the <a href="https://arxiv.org/abs/1712.07629v4">original paper.</a> </small>

## Usage tips

Here is a quick example of using the model to detect interest points in an image:

```python
from transformers import AutoImageProcessor, AutoModel
from transformers import AutoImageProcessor, SuperPointForKeypointDetection
import torch
from PIL import Image
import requests
Expand All @@ -52,7 +57,7 @@ url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)

processor = AutoImageProcessor.from_pretrained("magic-leap-community/superpoint")
model = AutoModel.from_pretrained("magic-leap-community/superpoint")
model = SuperPointForKeypointDetection.from_pretrained("magic-leap-community/superpoint")

inputs = processor(image, return_tensors="pt")
outputs = model(**inputs)
Expand All @@ -64,7 +69,7 @@ You can also feed multiple images to the model. Due to the nature of SuperPoint,
you will need to use the mask attribute to retrieve the respective information :

```python
from transformers import AutoImageProcessor, AutoModel
from transformers import AutoImageProcessor, SuperPointForKeypointDetection
import torch
from PIL import Image
import requests
Expand All @@ -77,7 +82,7 @@ image_2 = Image.open(requests.get(url_image_2, stream=True).raw)
images = [image_1, image_2]

processor = AutoImageProcessor.from_pretrained("magic-leap-community/superpoint")
model = AutoModel.from_pretrained("magic-leap-community/superpoint")
model = SuperPointForKeypointDetection.from_pretrained("magic-leap-community/superpoint")

inputs = processor(images, return_tensors="pt")
outputs = model(**inputs)
Expand All @@ -103,6 +108,12 @@ cv2.imwrite("output_image.png", image)
This model was contributed by [stevenbucaille](https://huggingface.co/stevenbucaille).
The original code can be found [here](https://github.com/magicleap/SuperPointPretrainedNetwork).

## Resources

A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with SuperPoint. If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.

- A notebook showcasing inference and visualization with SuperPoint can be found [here](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/SuperPoint/Inference_with_SuperPoint_to_detect_interest_points_in_an_image.ipynb). 🌎

## SuperPointConfig

[[autodoc]] SuperPointConfig
Expand Down
Loading