This repository is a curated collection of research papers, products, and datasets focused on video generation and editing. It serves as a comprehensive resource for researchers and practitioners in the field, providing organized links to studies, code, and benchmarks, with a particular emphasis on diffusion models.

How It Works

The repository categorizes advancements in video generation by task (text-to-video, image-to-video, audio-to-video, personalized generation, video editing) and publication venue/year. It links directly to research papers (often via arXiv), associated code repositories, project pages, and sometimes demos or benchmarks, facilitating easy access to state-of-the-art techniques.

Quick Start & Requirements

This is a curated list of research papers and resources, not a runnable software package. No installation or specific requirements are needed to browse the content.

Highlighted Details

• Extensive coverage of Text-to-Video generation, including major conferences like CVPR, ECCV, ICLR, and AAAI, with papers dating back to 2021.
• Detailed sections on Image-to-Video, Personalized Video Generation, and Video Editing, showcasing recent advancements and techniques.
• Includes a dedicated section for relevant datasets (e.g., UCF101, DAVIS, WebVid-10M, InternVid) and evaluation metrics (e.g., VBench, FVD).
• Features a "Products" table summarizing key commercial and research video generation models like Sora, Lumiere, and VideoPoet.

Maintenance & Community

The repository is actively maintained, with recent updates noted for NeurIPS 2023, AAAI 2024, and ECCV 2024 papers. Users are encouraged to contribute via issues and pull requests.

Licensing & Compatibility

The repository itself is not software and thus has no direct licensing implications. Individual linked papers and code repositories will have their own licenses.

Limitations & Caveats

This is a curated list and does not provide direct access to runnable code or models. The "Products" section lists commercial offerings, not open-source implementations.