<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> This repository serves as a comprehensive, curated list of research papers focused on deep learning-based video compression. It is designed for researchers, engineers, and power users seeking to understand the rapidly evolving landscape of AI-driven video compression techniques. By organizing papers thematically and chronologically, it provides a valuable resource for identifying state-of-the-art methods, emerging trends, and key contributions in the field, enabling informed decisions about adopting, researching, or building upon these technologies.

How It Works

The repository itself is a catalog of academic work, not an executable system. The underlying research explores various deep learning architectures and methodologies to achieve superior video compression performance. These approaches often involve end-to-end learned models that optimize for rate-distortion trade-offs, perceptual quality, and computational efficiency. Techniques highlighted include generative adversarial networks (GANs), implicit neural representations, transformer-based models, and advanced motion estimation/prediction strategies, aiming to surpass traditional video coding standards by leveraging the power of neural networks for complex data modeling.

Quick Start & Requirements

This repository is a curated list of research papers and does not contain any software to install or run. Therefore, a "Quick Start" or "Requirements" section is not applicable.

Highlighted Details

• Extensive Categorization: Papers are meticulously organized into thematic categories such as Generative Compression, Implicit Neural Representation, Video Coding for Machines, and Rate Control, alongside chronological grouping by publication year.
• Chronological Scope: The list covers research from 2017 to recent publications, including pre-prints and papers slated for 2025, offering a broad historical perspective and tracking the field's progression.
• Focus on Current Research: The repository emphasizes recent advancements, particularly "After June 2024," highlighting cutting-edge developments.
• Inclusion of Surveys and Benchmarks: The "Benchmark & Dataset & Survey" section points to resources that aid in evaluating and understanding the current state and future directions of deep learning-based video compression.

Maintenance & Community

The provided README does not contain information regarding project maintenance, active contributors, community channels (e.g., Discord, Slack), or a public roadmap. It appears to be a static compilation of research papers.

Licensing & Compatibility

No licensing information is provided within the README. This absence prevents an assessment of compatibility for commercial use, integration into proprietary systems, or other licensing restrictions.

Limitations & Caveats

This repository is a bibliographic resource and does not provide any code, implementations, or executables. Users cannot directly use or test any of the described methods from this repository alone. Furthermore, the lack of licensing details is a significant barrier to understanding potential usage rights and restrictions.