This repository serves as a curated digest of recent advancements in collaborative, cooperative, and multi-agent perception for V2I/V2V/V2X autonomous driving scenarios. It targets researchers and engineers in the autonomous driving domain, providing a structured overview of methods, datasets, and simulators.

How It Works

The project compiles and categorizes research papers, offering links to their respective resources. It focuses on the methodologies, frameworks, and datasets used in collaborative perception, aiming to track the evolution of techniques in this field. The curator notes potential inconsistencies in benchmark results across papers due to varying evaluation settings.

Quick Start & Requirements

This repository is a paper digest and does not have a direct installation or execution command. It links to various external libraries and datasets, each with its own requirements.

Highlighted Details

• Extensive collection of papers categorized by conference (CVPR, ECCV, NeurIPS, etc.) and year.
• Provides links to associated code, datasets, and project pages for many entries.
• Includes a comprehensive list of recommended surveys, talks, libraries, and influential researchers in the field.
• Organized into "Method and Framework" and "Dataset and Simulator" sections for easy navigation.

Maintenance & Community

The repository is maintained by Little-Podi and lists several influential researchers in the field. It links to various workshops and external libraries like V2Xverse, HEAL, and OpenCOOD.

Licensing & Compatibility

The repository itself does not specify a license. Individual linked papers and code repositories will have their own licenses, which may vary.

Limitations & Caveats

The curator explicitly states that direct comparison of benchmark results across papers is difficult due to inconsistent evaluation settings. Some linked code repositories are marked as unavailable (~~code~~).