ICCV 2023-25 Papers - Release Candidate (v2025.10.27-rc.1)

ICCV 2023-25 Papers - Release Candidate (v2025.10.27-rc.1)

---

Overview

Dive into the latest cutting-edge research from the leading computer vision conference - ICCV 2025.

Each entry may include links to available resources:

• Paper on thecvf / arXiv
• Code on GitHub
• Models or demos on Hugging Face
• Supplementary media (e.g., YouTube)

Not all papers provide every resource - only the available badges are shown.

---

Highlights

Introduced new sections for ICCV 2025 Papers:

Orals

• Multi-Modal Learning
• Structure and Motion
• Efficient Learning
• View Synthesis and Scene Reconstruction
• Foundation Models and Representation Learning
• Human Modeling
• 3D Pose Understanding
• Vision + Graphics
• Applications and Evaluation
• Content Generation
• Physical Scene Perception
• Segmentation and Grouping

---

Full Changelog: https://github.com/DmitryRyumin/ICCV-2023-25-Papers/commits/v2025.10.27-rc.1

---

Happy Researching!