PVUW 2025 Challenge Report: Advances in Pixel-level Understanding of Complex Videos in the Wild
This is an incremental challenge report for researchers in computer vision, focusing on advancing pixel-level video understanding in real-world scenarios.
The report summarizes the 4th PVUW Challenge at CVPR 2025, which tackled complex video segmentation through two tracks (MOSE and MeViS) with new challenging datasets, providing insights into state-of-the-art methods and trends.
This report provides a comprehensive overview of the 4th Pixel-level Video Understanding in the Wild (PVUW) Challenge, held in conjunction with CVPR 2025. It summarizes the challenge outcomes, participating methodologies, and future research directions. The challenge features two tracks: MOSE, which focuses on complex scene video object segmentation, and MeViS, which targets motion-guided, language-based video segmentation. Both tracks introduce new, more challenging datasets designed to better reflect real-world scenarios. Through detailed evaluation and analysis, the challenge offers valuable insights into the current state-of-the-art and emerging trends in complex video segmentation. More information can be found on the workshop website: https://pvuw.github.io/.