Repository files navigation

TaskMatrix connects ChatGPT and a series of Visual Foundation Models to enablesending andreceiving images during chatting.

See our paper:Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models

• Now TaskMatrix supportsGroundingDINO andsegment-anything! Thanks@jordddan for his efforts. For the image editing case,GroundingDINO is first used to locate bounding boxes guided by given text, thensegment-anything is used to generate the related mask, and finally stable diffusion inpainting is used to edit image based on the mask.

  • Firstly, runpython visual_chatgpt.py --load "Text2Box_cuda:0,Segmenting_cuda:0,Inpainting_cuda:0,ImageCaptioning_cuda:0"
  • Then, sayfind xxx in the image orsegment xxx in the image.xxx is an object. TaskMatrix will return the detection or segmentation result!

• Now TaskMatrix can support Chinese! Thanks to@Wang-Xiaodong1899 for his efforts.

• We propose thetemplate idea in TaskMatrix!

  • A template is apre-defined execution flow that assists ChatGPT in assembling complex tasks involving multiple foundation models.
  • A template contains theexperiential solution to complex tasks as determined by humans.
  • A template caninvoke multiple foundation models or evenestablish a new ChatGPT session
  • To define atemplate, simply adding a class with attributestemplate_model = True

• Thanks to@ShengmingYin and@thebestannie for providing a template example inInfinityOutPainting class (see the following gif)

  • Firstly, runpython visual_chatgpt.py --load "Inpainting_cuda:0,ImageCaptioning_cuda:0,VisualQuestionAnswering_cuda:0"
  • Secondly, sayextend the image to 2048x1024 to TaskMatrix!
  • By simply creating anInfinityOutPainting template, TaskMatrix can seamlessly extend images to any size through collaboration with existingImageCaptioning,Inpainting, andVisualQuestionAnswering foundation models,without the need for additional training.

• TaskMatrix needs the effort of the community! We crave your contribution to add new and interesting features!

On the one hand,ChatGPT (or LLMs) serves as ageneral interface that provides a broad and diverse understanding of awide range of topics. On the other hand,Foundation Models serve asdomain experts by providing deep knowledge in specific domains.By leveragingboth general and deep knowledge, we aim at building an AI that is capable of handling various tasks.

# clone the repogit clone https://github.com/microsoft/TaskMatrix.git# Go to directorycd visual-chatgpt# create a new environmentconda create -n visgpt python=3.8# activate the new environmentconda activate visgpt#  prepare the basic environmentspip install -r requirements.txtpip install  git+https://github.com/IDEA-Research/GroundingDINO.gitpip install  git+https://github.com/facebookresearch/segment-anything.git# prepare your private OpenAI key (for Linux)export OPENAI_API_KEY={Your_Private_Openai_Key}# prepare your private OpenAI key (for Windows)set OPENAI_API_KEY={Your_Private_Openai_Key}# Start TaskMatrix !# You can specify the GPU/CPU assignment by "--load", the parameter indicates which # Visual Foundation Model to use and where it will be loaded to# The model and device are separated by underline '_', the different models are separated by comma ','# The available Visual Foundation Models can be found in the following table# For example, if you want to load ImageCaptioning to cpu and Text2Image to cuda:0# You can use: "ImageCaptioning_cpu,Text2Image_cuda:0"# Advice for CPU Userspython visual_chatgpt.py --load ImageCaptioning_cpu,Text2Image_cpu# Advice for 1 Tesla T4 15GB  (Google Colab)                       python visual_chatgpt.py --load "ImageCaptioning_cuda:0,Text2Image_cuda:0"                                # Advice for 4 Tesla V100 32GB                            python visual_chatgpt.py --load "Text2Box_cuda:0,Segmenting_cuda:0,    Inpainting_cuda:0,ImageCaptioning_cuda:0,    Text2Image_cuda:1,Image2Canny_cpu,CannyText2Image_cuda:1,    Image2Depth_cpu,DepthText2Image_cuda:1,VisualQuestionAnswering_cuda:2,    InstructPix2Pix_cuda:2,Image2Scribble_cpu,ScribbleText2Image_cuda:2,    SegText2Image_cuda:2,Image2Pose_cpu,PoseText2Image_cuda:2,    Image2Hed_cpu,HedText2Image_cuda:3,Image2Normal_cpu,    NormalText2Image_cuda:3,Image2Line_cpu,LineText2Image_cuda:3"

Here we list the GPU memory usage of each visual foundation model, you can specify which one you like:

We appreciate the open source of the following projects:

Hugging Face  LangChain  Stable Diffusion  ControlNet  InstructPix2Pix  CLIPSeg  BLIP  

For help or issues using the TaskMatrix, please submit a GitHub issue.

For other communications, please contact Chenfei WU (chewu@microsoft.com) or Nan DUAN (nanduan@microsoft.com).

Trademarks This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must followMicrosoft’s Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party’s policies.