# VTimeLLM \[[Paper](https://arxiv.org/pdf/2311.18445.pdf)\] Official PyTorch implementation of the paper "VTimeLLM: Empower LLM to Grasp Video Moments". [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/vtimellm-empower-llm-to-grasp-video-moments/video-based-generative-performance-4)](https://paperswithcode.com/sota/video-based-generative-performance-4?p=vtimellm-empower-llm-to-grasp-video-moments) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/vtimellm-empower-llm-to-grasp-video-moments/dense-video-captioning-on-activitynet)](https://paperswithcode.com/sota/dense-video-captioning-on-activitynet?p=vtimellm-empower-llm-to-grasp-video-moments) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/vtimellm-empower-llm-to-grasp-video-moments/video-based-generative-performance-3)](https://paperswithcode.com/sota/video-based-generative-performance-3?p=vtimellm-empower-llm-to-grasp-video-moments) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/vtimellm-empower-llm-to-grasp-video-moments/video-based-generative-performance-1)](https://paperswithcode.com/sota/video-based-generative-performance-1?p=vtimellm-empower-llm-to-grasp-video-moments) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/vtimellm-empower-llm-to-grasp-video-moments/video-based-generative-performance-5)](https://paperswithcode.com/sota/video-based-generative-performance-5?p=vtimellm-empower-llm-to-grasp-video-moments) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/vtimellm-empower-llm-to-grasp-video-moments/video-based-generative-performance-2)](https://paperswithcode.com/sota/video-based-generative-performance-2?p=vtimellm-empower-llm-to-grasp-video-moments) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/vtimellm-empower-llm-to-grasp-video-moments/video-based-generative-performance)](https://paperswithcode.com/sota/video-based-generative-performance?p=vtimellm-empower-llm-to-grasp-video-moments) --- ## :loudspeaker: Latest Updates - **Jan-2**: Thanks to [Xiao Xia](https://github.com/Rishubi) , [Shengbo Tong](https://github.com/tsb-19) and [Beining Wang](https://github.com/Benson0704), we have refactored the code to now support both the LLAMA and ChatGLM3 architectures. We translated the training data into Chinese and fine-tuned a Chinese version based on the ChatGLM3-6b. - **Dec-14**: Released the training code and data. All the resources including models, datasets and extracted features are available [here](https://cloud.tsinghua.edu.cn/d/6db5d02883124826aa6f/?p=%2F&mode=list). :fire::fire: - **Dec-4**: VTimeLLM: demo released. --- ## VTimeLLM Overview :bulb: VTimeLLM is a novel Video LLM designed for fine-grained video moment understanding and reasoning with respect to time boundary. VTimeLLM adopts a boundary-aware three-stage training strategy, which respectively utilizes image-text pairs for feature alignment, multiple-event videos to increase temporal-boundary awareness, and high-quality video-instruction tuning to further improve temporal understanding ability as well as align with human intents. ![framework](images/framework.png) --- ## Contributions :trophy: - We propose VTimeLLM, the first boundary-aware Video LLM, to the best of our knowledge. - We propose the boundary-aware three-stage training strategy, which consecutively leverages i) large-scale image-text data for feature alignment, ii) large-scale multi-event video-text data together with the temporal-related single-turn and multi-turn QA to enhance the awareness of time boundary, and iii) instruction tuning on the high-quality dialog dataset for better temporal reasoning ability. - We conduct extensive experiments to demonstrate that the proposed VTimeLLM significantly outperforms existing Video LLMs in various fine-grained temporal-related video tasks, showing its superior ability for video understanding and reasoning. --- ## Installation :wrench: We recommend setting up a conda environment for the project: ```shell conda create --name=vtimellm python=3.10 conda activate vtimellm git clone https://github.com/huangb23/VTimeLLM.git cd VTimeLLM pip install -r requirements.txt ``` Additionally, install additional packages for training cases. ```shell pip install ninja pip install flash-attn --no-build-isolation ``` ## Running Demo Offline :cd: To run the demo offline, please refer to the instructions in [offline_demo.md](docs/offline_demo.md). ## Training :train: For training instructions, check out [train.md](docs/train.md). ## Qualitative Analysis :mag: A Comprehensive Evaluation of VTimeLLM's Performance across Multiple Tasks. ### Video Understanding and Conversational Tasks :speech_balloon: ![0](images/ex.png) --- ### Creative Tasks :paintbrush: ![1](images/ex1.png) --- ### Fine-grained Understanding Tasks :globe_with_meridians: ![2](images/ex2.png) --- ### Video Reasoning Tasks :question: ![3](images/ex3.png) --- ## Acknowledgements :pray: We are grateful for the following awesome projects our VTimeLLM arising from: * [LLaVA](https://github.com/haotian-liu/LLaVA): Large Language and Vision Assistant * [FastChat](https://github.com/lm-sys/FastChat): An Open Platform for Training, Serving, and Evaluating Large Language Model based Chatbots * [Video-ChatGPT](https://github.com/mbzuai-oryx/Video-ChatGPT): Towards Detailed Video Understanding via Large Vision and Language Models * [LLaMA](https://github.com/facebookresearch/llama): Open and Efficient Foundation Language Models * [Vid2seq](https://github.com/google-research/scenic/tree/main/scenic/projects/vid2seq): Large-Scale Pretraining of a Visual Language Model for Dense Video Captioning * [InternVid](https://github.com/OpenGVLab/InternVideo/tree/main/Data/InternVid): A Large-scale Video-Text dataset If you're using VTimeLLM in your research or applications, please cite using this BibTeX: ```bibtex @inproceedings{huang2024vtimellm, title={Vtimellm: Empower llm to grasp video moments}, author={Huang, Bin and Wang, Xin and Chen, Hong and Song, Zihan and Zhu, Wenwu}, booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition}, pages={14271--14280}, year={2024} } ``` ## License :scroll: Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License. Looking forward to your feedback, contributions, and stars! :star2: