Ask-Anything

中文 README

Currently, Ask-Anything is a simple yet interesting tool for chatting with video. Our team is trying to build a smart and robust chatbot that can understand video.

We are also working on a updated version, stay tuned! ⭐️

:movie_camera: Online Demo [click here]

https://user-images.githubusercontent.com/24236723/233630363-b20304ab-763b-40e5-b526-e2a6b9e9cae2.mp4

:fire: Updates

• 2023/04/25 Watch videos longer than one minute with chatGPT

  • VideoChat_LongVideo: Update langchain to the latest version.

• 2023/04/21 Chat with MOSS

  • video_chat_with_MOSS: Explicit communication with MOSS.

• 2023/04/20: Chat with StableLM

  • video_chat_with_StableLM: Explicit communication with StableLM.

• 2023/04/19: Code release & Online Demo

  • VideoChat: Explicit communication with ChatGPT. Sensitive with time. demo is avaliable!
  • MiniGPT-4 for video: Implicit communication with Vicuna. Not sensitive with time. (Simple extension of MiniGPT-4, which will be improved in the future.)

:speech_balloon: Example

https://user-images.githubusercontent.com/24236723/233631602-6a69d83c-83ef-41ed-a494-8e0d0ca7c1c8.mp4

🔨 Getting Started

Build video chat with:

• ChatGPT
• StableLM
• MiniGPT-4
• MOSS

:hourglass_flowing_sand: Ongoing

Our team constantly studies general video understanding and long-term video reasoning:

• [ ] Strong video foundation model.
• [ ] Video-text dataset and video reasoning benchmark.
• [ ] Video-language system with LLMs.
• [ ] Artificial Intelligence Generated Content (AIGC) for Video.
• [ ] ...

We are hiring researchers, engineers and interns in General Vision Group, Shanghai AI Lab. If you are interested in working with us, please contact Yi Wang ([email protected]).