/Grounded-Segment-Anything

Marrying Grounding DINO with Segment Anything & Stable Diffusion & Tag2Text & BLIP & Whisper & ChatBot - Automatically Detect , Segment and Generate Anything with Image, Text, and Audio Inputs

Primary LanguageJupyter NotebookApache License 2.0Apache-2.0

Grounded-Segment-Anything

YouTube Colab Open in Colab HuggingFace Space ModelScope Official Demo Huggingface Demo by Community Stable-Diffusion WebUI Jupyter Notebook Demo

We plan to create a very interesting demo by combining Grounding DINO and Segment Anything which aims to detect and segment Anything with text inputs! And we will continue to improve it and create more interesting demos based on this foundation.

We are very willing to help everyone share and promote new projects based on Segment-Anything, Please checkout here for more amazing demos and works in the community: Highlight Extension Projects. You can submit a new issue (with project tag) or a new pull request to add new project's links.

🍄 Why Building this Project?

The core idea behind this project is to combine the strengths of different models in order to build a very powerful pipeline for solving complex problems. And it's worth mentioning that this is a workflow for combining strong expert models, where all parts can be used separately or in combination, and can be replaced with any similar but different models (like replacing Grounding DINO with GLIP or other detectors / replacing Stable-Diffusion with ControlNet or GLIGEN/ Combining with ChatGPT).

🍊 Preliminary Works

  • Segment Anything is a strong segmentation model. But it needs prompts (like boxes/points) to generate masks.
  • Grounding DINO is a strong zero-shot detector which is capable of to generate high quality boxes and labels with free-form text.
  • OSX is a strong and efficient one-stage motion capture method to generate high quality 3D human mesh from monucular image. We also release a large-scale upper-body dataset UBody for a more accurate reconstrution in the upper-body scene.
  • Stable-Diffusion is an amazing strong text-to-image diffusion model.
  • Tag2Text is an efficient and controllable vision-language model which can simultaneously output superior image captioning and image tagging.
  • BLIP is a wonderful language-vision model for image understanding.
  • Visual ChatGPT is a wonderful tool that connects ChatGPT and a series of Visual Foundation Models to enable sending and receiving images during chatting.
  • VoxelNeXt is a clean, simple, and fully-sparse 3D object detector, which predicts objects directly upon sparse voxel features.

🔥 Highlighted Projects

🍉 The Supported Amazing Demos in this Project

The Amazing Demo Preview (Continual Updating)

🔥 ChatBot for our project is built

chatbot.mp4

🔥 🔈Speak to edit🎨: Whisper + ChatGPT + Grounded-SAM + SD

🔥 Grounded-SAM: Semi-automatic Labeling System

Tips

  • If you want to detect multiple objects in one sentence with Grounding DINO, we suggest seperating each name with . . An example: cat . dog . chair .

🔥 Grounded-SAM + Stable-Diffusion Inpainting: Data-Factory, Generating New Data

🔥 Tag2Text + Grounded-SAM: Automatic Label System with Superior Image Tagging

Using Tag2Text to directly generate tags, and using Grounded-SAM for box and mask generating. Tag2Text has superior tagging and captioning capabilities. Here's the demo output comparison:

🔥 BLIP + Grounded-SAM: Automatic Label System

Using BLIP to generate caption, extracting tags with ChatGPT, and using Grounded-SAM for box and mask generating. Here's the demo output:

🔥 Grounded-SAM+OSX: Promptable 3D Whole-Body Human Mesh Recovery

Using Grounded-SAM for box and mask generating, using OSX to estimate the SMPLX parameters and reconstruct 3D whole-body (body, face and hand) human mesh. Here's a demo:


🔥 Interactive Editing

  • Release the interactive fashion-edit playground in here. Run in the notebook, just click for annotating points for further segmentation. Enjoy it!

  • Release human-face-edit branch here. We'll keep updating this branch with more interesting features. Here are some examples:

🔥 3D-Box via Segment Anything We extend the scope to 3D world by combining Segment Anything and VoxelNeXt. When we provide a prompt (e.g., a point / box), the result is not only 2D segmentation mask, but also 3D boxes.

💡 Highlight Extension Projects

📖 Notebook Demo

See our notebook file as an example.

🛠️ Installation

The code requires python>=3.8, as well as pytorch>=1.7 and torchvision>=0.8. Please follow the instructions here to install both PyTorch and TorchVision dependencies. Installing both PyTorch and TorchVision with CUDA support is strongly recommended.

Install with Docker

Open one terminal:

make run

That's it.

If you would like to allow visualization across docker container, open another terminal and type:

xhost +

Install without Docker

You should set the environment variable manually as follows if you want to build a local GPU environment for Grounded-SAM:

export AM_I_DOCKER=False
export BUILD_WITH_CUDA=True
export CUDA_HOME=/path/to/cuda-11.3/

Install Segment Anything:

python -m pip install -e segment_anything

Install Grounding DINO:

python -m pip install -e GroundingDINO

Install diffusers:

pip install --upgrade diffusers[torch]

Install osx:

git submodule update --init --recursive
cd grounded-sam-osx && bash install.sh

Install Tag2Text:

git submodule update --init --recursive
cd Tag2Text && pip install -r requirements.txt

The following optional dependencies are necessary for mask post-processing, saving masks in COCO format, the example notebooks, and exporting the model in ONNX format. jupyter is also required to run the example notebooks.

pip install opencv-python pycocotools matplotlib onnxruntime onnx ipykernel

More details can be found in install segment anything and install GroundingDINO and install OSX

🏃 Run Grounding DINO Demo

  • Download the checkpoint for Grounding Dino:
cd Grounded-Segment-Anything

wget https://github.com/IDEA-Research/GroundingDINO/releases/download/v0.1.0-alpha/groundingdino_swint_ogc.pth
  • Run demo
export CUDA_VISIBLE_DEVICES=0
python grounding_dino_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --input_image assets/demo1.jpg \
  --output_dir "outputs" \
  --box_threshold 0.3 \
  --text_threshold 0.25 \
  --text_prompt "bear" \
  --device "cuda"
  • The model prediction visualization will be saved in output_dir as follow:

🏃‍♂️ Run Grounded-Segment-Anything Demo

  • Download the checkpoint for Segment Anything and Grounding Dino:
cd Grounded-Segment-Anything

wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth
wget https://github.com/IDEA-Research/GroundingDINO/releases/download/v0.1.0-alpha/groundingdino_swint_ogc.pth
  • Run Demo
export CUDA_VISIBLE_DEVICES=0
python grounded_sam_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --input_image assets/demo1.jpg \
  --output_dir "outputs" \
  --box_threshold 0.3 \
  --text_threshold 0.25 \
  --text_prompt "bear" \
  --device "cuda"
  • The model prediction visualization will be saved in output_dir as follow:

More Examples

⛷️ Run Grounded-Segment-Anything + Inpainting Demo

CUDA_VISIBLE_DEVICES=0
python grounded_sam_inpainting_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --input_image assets/inpaint_demo.jpg \
  --output_dir "outputs" \
  --box_threshold 0.3 \
  --text_threshold 0.25 \
  --det_prompt "bench" \
  --inpaint_prompt "A sofa, high quality, detailed" \
  --device "cuda"

🏌️ Run Grounded-Segment-Anything + Inpainting Gradio APP

The following introduces the 6 task_type modes provided by Gradio APP:

  1. scribble: Segmentation is achieved through Segment Anything and mouse click interaction (you need to click on the object with the mouse, no need to specify the prompt).
  2. automask: Segment the entire image at once through Segment Anything (no need to specify a prompt).
  3. det: Realize detection through Grounding DINO and text interaction (text prompt needs to be specified).
  4. seg: Realize text interaction by combining Grounding DINO and Segment Anything to realize detection + segmentation (need to specify text prompt).
  5. inpainting: By combining Grounding DINO + Segment Anything + Stable Diffusion to achieve text exchange and replace the target object (need to specify text prompt and inpaint prompt) .
  6. automatic: By combining BLIP + Grounding DINO + Segment Anything to achieve non-interactive detection + segmentation (no need to specify prompt).
python gradio_app.py
  • The gradio_app visualization as follow:

🏷️ Run Grounded-Segment-Anything + Tag2Text Demo

Tag2Text achieves superior image tag recognition ability of 3,429 commonly human-used categories. It is seamlessly linked to generate pseudo labels automatically as follows:

  1. Use Tag2Text to generate tags.
  2. Use Grounded-Segment-Anything to generate the boxes and masks.
  • Download the checkpoint for Tag2Text:
cd Tag2Text

wget https://huggingface.co/spaces/xinyu1205/Tag2Text/resolve/main/tag2text_swin_14m.pth
  • Run Demo
export CUDA_VISIBLE_DEVICES=0
python automatic_label_tag2text_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --tag2text_checkpoint tag2text_swin_14m.pth \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --input_image assets/demo9.jpg \
  --output_dir "outputs" \
  --box_threshold 0.25 \
  --text_threshold 0.2 \
  --iou_threshold 0.5 \
  --device "cuda"
  • Tag2Text also provides powerful captioning capabilities, and the process with captions can refer to BLIP.
  • The pseudo labels and model prediction visualization will be saved in output_dir as follows (right figure):

🤖 Run Grounded-Segment-Anything + BLIP Demo

It is easy to generate pseudo labels automatically as follows:

  1. Use BLIP (or other caption models) to generate a caption.
  2. Extract tags from the caption. We use ChatGPT to handle the potential complicated sentences.
  3. Use Grounded-Segment-Anything to generate the boxes and masks.
  • Run Demo
export OPENAI_API_KEY=your_openai_key
export OPENAI_API_BASE=https://closeai.deno.dev/v1
export CUDA_VISIBLE_DEVICES=0
python automatic_label_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --input_image assets/demo3.jpg \
  --output_dir "outputs" \
  --openai_key $OPENAI_API_KEY \
  --box_threshold 0.25 \
  --text_threshold 0.2 \
  --iou_threshold 0.5 \
  --device "cuda"
  • When you don't have a paid Account for ChatGPT is also possible to use NLTK instead. Just don't include the openai_key Parameter when starting the Demo.
    • The Script will automatically download the necessary NLTK Data.
  • The pseudo labels and model prediction visualization will be saved in output_dir as follows:

😮 Run Grounded-Segment-Anything + Whisper Demo

Detect and segment anything with speech!

Install Whisper

pip install -U openai-whisper

See the whisper official page if you have other questions for the installation.

Run Voice-to-Label Demo

Optional: Download the demo audio file

wget https://huggingface.co/ShilongLiu/GroundingDINO/resolve/main/demo_audio.mp3
export CUDA_VISIBLE_DEVICES=0
python grounded_sam_whisper_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --input_image assets/demo4.jpg \
  --output_dir "outputs" \
  --box_threshold 0.3 \
  --text_threshold 0.25 \
  --speech_file "demo_audio.mp3" \
  --device "cuda"

Run Voice-to-inpaint Demo

You can enable chatgpt to help you automatically detect the object and inpainting order with --enable_chatgpt.

Or you can specify the object you want to inpaint [stored in args.det_speech_file] and the text you want to inpaint with [stored in args.inpaint_speech_file].

export OPENAI_API_KEY=your_openai_key
export OPENAI_API_BASE=https://closeai.deno.dev/v1
# Example: enable chatgpt
export CUDA_VISIBLE_DEVICES=0
python grounded_sam_whisper_inpainting_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --input_image assets/inpaint_demo.jpg \
  --output_dir "outputs" \
  --box_threshold 0.3 \
  --text_threshold 0.25 \
  --prompt_speech_file assets/acoustics/prompt_speech_file.mp3 \
  --enable_chatgpt \
  --openai_key $OPENAI_API_KEY\
  --device "cuda"
# Example: without chatgpt
export CUDA_VISIBLE_DEVICES=0
python grounded_sam_whisper_inpainting_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --input_image assets/inpaint_demo.jpg \
  --output_dir "outputs" \
  --box_threshold 0.3 \
  --text_threshold 0.25 \
  --det_speech_file "assets/acoustics/det_voice.mp3" \
  --inpaint_speech_file "assets/acoustics/inpaint_voice.mp3" \
  --device "cuda"

💬 Run ChatBot Demo

Following Visual ChatGPT, we add a ChatBot for our project. Currently, it supports:

  1. "Descripe the image."
  2. "Detect the dog (and the cat) in the image."
  3. "Segment anything in the image."
  4. "Segment the dog (and the cat) in the image."
  5. "Help me label the image."
  6. "Replace the dog with a cat in the image."

To use the ChatBot:

  • Install whisper if you want to use audio as input.
  • Set the default model setting in the tool Grounded_dino_sam_inpainting.
  • Run Demo
export OPENAI_API_KEY=your_openai_key
export OPENAI_API_BASE=https://closeai.deno.dev/v1
export CUDA_VISIBLE_DEVICES=0
python chatbot.py

🕺 Run Grounded-Segment-Anything + OSX Demo

  • Download the checkpoint osx_l_wo_decoder.pth.tar from here for OSX:

  • Download the human model files and place it into grounded-sam-osx/utils/human_model_files following the instruction of OSX.

  • Run Demo

export CUDA_VISIBLE_DEVICES=0
python grounded_sam_osx_demo.py \
  --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \
  --grounded_checkpoint groundingdino_swint_ogc.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --osx_checkpoint osx_l_wo_decoder.pth.tar \
  --input_image assets/osx/grounded_sam_osx_demo.png \
  --output_dir "outputs" \
  --box_threshold 0.3 \
  --text_threshold 0.25 \
  --text_prompt "humans, chairs" \
  --device "cuda"
  • The model prediction visualization will be saved in output_dir as follow:

  • We also support promptable 3D whole-body mesh recovery. For example, you can track someone with with a text prompt and estimate his 3D pose and shape :
space-1.jpg
A person with pink clothes
space-1.jpg
A man with a sunglasses

🕺 Run Grounded-Segment-Anything + VISAM Demo

  • Download the checkpoint motrv2_dancetrack.pth from here for MOTRv2:

  • See the more thing if you have other questions for the installation.

  • Run Demo

export CUDA_VISIBLE_DEVICES=0
python grounded_sam_visam.py \
  --meta_arch motr \
  --dataset_file e2e_dance \
  --with_box_refine \
  --query_interaction_layer QIMv2 \
  --num_queries 10 \
  --det_db det_db_motrv2.json \
  --use_checkpoint \
  --mot_path your_data_path \
  --resume motrv2_dancetrack.pth \
  --sam_checkpoint sam_vit_h_4b8939.pth \
  --video_path DanceTrack/test/dancetrack0003

||

💘 Acknowledgements

Contributors

Our project wouldn't be possible without the contributions of these amazing people! Thank you all for making this project better.

Citation

If you find this project helpful for your research, please consider citing the following BibTeX entry.

@article{kirillov2023segany,
  title={Segment Anything}, 
  author={Kirillov, Alexander and Mintun, Eric and Ravi, Nikhila and Mao, Hanzi and Rolland, Chloe and Gustafson, Laura and Xiao, Tete and Whitehead, Spencer and Berg, Alexander C. and Lo, Wan-Yen and Doll{\'a}r, Piotr and Girshick, Ross},
  journal={arXiv:2304.02643},
  year={2023}
}

@inproceedings{ShilongLiu2023GroundingDM,
  title={Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection},
  author={Shilong Liu and Zhaoyang Zeng and Tianhe Ren and Feng Li and Hao Zhang and Jie Yang and Chunyuan Li and Jianwei Yang and Hang Su and Jun Zhu and Lei Zhang},
  year={2023}
}