STAC: Selective Spatiotemporal Aggregation and Compression for Video Reasoning Segmentation

Benchmark Model Rank Results
referring-video-object-segmentation-on-refvos-davisSTAC#3J&F: 73.6
referring-video-object-segmentation-on-reasonvosSTAC#9J&F: 52.3
referring-video-object-segmentation-on-mevisSTAC#15J&F: 49.9
referring-video-object-segmentation-on-revosSTAC#22J&F: 48.4
referring-expression-segmentation-on-refer-youtube-vos-2021-public-validationSTAC#26J&F: 65.2