This repository serves as a curated collection of papers and associated code for the ICCV 2025 conference, targeting researchers and practitioners in computer vision. It aims to provide a centralized, organized resource for the latest advancements and their implementations, facilitating easier access to cutting-edge research.

How It Works

The repository categorizes ICCV 2025 papers by research topic, ranging from foundational areas like object detection and segmentation to emerging fields such as 3D Gaussian Splatting, multimodal large language models, and embodied AI. Each entry typically includes links to the paper's abstract (arXiv) and its corresponding code repository, enabling users to quickly find and explore relevant work.

Quick Start & Requirements

This repository is a curated list and does not require installation or execution. Users can browse the categorized links to access papers and code.

Highlighted Details

• Comprehensive coverage of ICCV 2025 papers, with an acceptance rate of 24% (2699/11239).
• Organized by over 50 specific computer vision sub-fields, including 3D Vision, Generative Models, Multimodal AI, and Autonomous Driving.
• Provides direct links to arXiv papers and GitHub repositories for each listed research item.
• Includes links to related resources for previous conferences (CVPR, ECCV) and a community knowledge base.

Maintenance & Community

The project encourages community contributions via GitHub issues to add new ICCV 2025 papers and code. It also promotes joining a large computer vision academic exchange group for real-time updates.

Licensing & Compatibility

Licensing information is not provided for the repository itself, as it is a collection of links. Individual paper and code licenses are determined by their respective authors and repositories.

Limitations & Caveats

The repository is a dynamic collection and relies on community submissions for completeness and accuracy. Links may become outdated, and the scope is limited to ICCV 2025 papers and their associated code.