UniPixel: Unified Object Referring and Segmentation for Pixel-Level Visual Reasoning

Benchmark Model Rank Results
referring-video-object-segmentation-on-mevisUniPixel (3B)–J&F: 53.1J: 50.4F: 55.7
referring-video-object-segmentation-on-mevisUniPixel (7B)–J&F: 55.8J: 53.2F: 58.3
referring-video-object-segmentation-on-revosUniPixel (3B)–J&F: 62.1J: 59.7F: 64.4
referring-video-object-segmentation-on-revosUniPixel (7B)–J&F: 63.7J: 61.7F: 65.7