|
--- |
|
tags: |
|
- vision-language model |
|
- llama |
|
- video understanding |
|
--- |
|
|
|
|
|
# LLaMA-VID Model Card |
|
<a href='https://llama-vid.github.io/'><img src='https://img.shields.io/badge/Project-Page-Green'></a> |
|
<a href='https://arxiv.org/abs/2311.17043'><img src='https://img.shields.io/badge/Paper-Arxiv-red'></a> |
|
|
|
## Model details |
|
LLaMA-VID empowers existing frameworks to support hour-long videos and pushes their upper limit with an extra context token. |
|
|
|
|
|
**Model type:** |
|
LLaMA-VID is an open-source chatbot trained by fine-tuning LLaMA/Vicuna on GPT-generated multimodal instruction-following data. |
|
LLaMA-VID empowers existing frameworks to support hour-long videos and pushes their upper limit with an extra context token. We build this repo based on LLaVA. |
|
|
|
|
|
**Model date:** |
|
llama-vid-7b-full-224-long-video was trained on 11/2023. |
|
|
|
## License |
|
Llama 2 is licensed under the LLAMA 2 Community License, |
|
Copyright (c) Meta Platforms, Inc. All Rights Reserved. |
|
|
|
**Where to send questions or comments about the model:** |
|
https://github.com/dvlab-research/LLaMA-VID/issues |
|
|
|
## Intended use |
|
**Primary intended uses:** |
|
The primary use of LLaMA-VID is research on large multimodal models and chatbots. |
|
|
|
**Primary intended users:** |
|
The primary intended users of the model are researchers and hobbyists in computer vision, natural language processing, machine learning, and artificial intelligence. |
|
|
|
## Training data |
|
This model is trained based on image data from LLaVA-1.5 dataset, and video data from WebVid and ActivityNet datasets, including |
|
- 558K filtered image-text pairs from LAION/CC/SBU, captioned by BLIP. |
|
- 158K GPT-generated multimodal instruction-following data. |
|
- 450K academic-task-oriented VQA data mixture. |
|
- 40K ShareGPT data. |
|
- 232K video-caption pairs sampled from the WebVid 2.5M dataset. |
|
- 98K videos from ActivityNet with QA pairs from Video-ChatGPT. |
|
- 15K video QA pairs from our Long VideoQA dataset. |
