This repository serves as a comprehensive, continually updated curated list of research works in Text-to-Video (T2V) and Image-to-Video (I2V) generation. It targets researchers, engineers, and practitioners seeking an organized overview of the rapidly evolving field, providing direct access to seminal papers, code, and datasets. The primary benefit is a centralized, well-structured resource for understanding the state-of-the-art and identifying relevant research directions.

How It Works

The project functions as a living bibliography, meticulously compiling and categorizing academic papers, associated code repositories, demos, and datasets. It is structured around key research areas like Text-to-Sequence-Image and Text-to-Video generation, offering detailed entries for each work. This approach allows users to quickly navigate and discover relevant advancements, leveraging the curated links for deeper exploration of specific methodologies and implementations.

Quick Start & Requirements

This repository is a curated list of research papers and resources, not a software package. Therefore, there are no installation or execution requirements.

Highlighted Details

• Features an extensive collection of over 100 Text-to-Video and Image-to-Video research papers, each linked to its respective publication, code, and demo where available.
• Includes a detailed compilation of 24 datasets and 9 evaluation metrics crucial for benchmarking video generation models.
• Organized into distinct sections covering Text-to-Sequence-Image and Text-to-Video research, facilitating targeted exploration.
• Based on the survey paper "From Sora What We Can See: A Survey of Text-to-Video Generation," providing a foundational overview.

Maintenance & Community

The repository is actively curated and maintained by Rui Sun and Yumin Zhang. While it references a survey paper and the GitHub repository itself, no specific community channels (e.g., Discord, Slack) or active contributor lists beyond the maintainers are detailed.

Licensing & Compatibility

The license for the curated list itself is not explicitly stated in the provided README. Individual research papers and their associated code repositories will have their own respective licenses, which users must consult separately. Compatibility for commercial use would depend on the licenses of the individual projects linked within the list.

Limitations & Caveats

As a curated list, its comprehensiveness is dependent on ongoing updates by the maintainers. It serves as a directory of research rather than a deployable tool for generating videos. Users must independently evaluate the licenses and applicability of individual linked projects.