URVOS: Unified Referring Video Object Segmentation Network with a Large-Scale Benchmark

Benchmark Model Rank Results
referring-expression-segmentation-on-davisURVOS + Refer-Youtube-VOS + ft. DAVIS#9J&F 1st frame: 51.63
referring-expression-segmentation-on-davisURVOS + Refer-Youtube-VOS#10J&F 1st frame: 46.85
referring-expression-segmentation-on-davisURVOS#14J&F 1st frame: 44.1
referring-expression-segmentation-on-refer-1URVOS#25J&F: 48.9J: 47.0F: 50.8
referring-video-object-segmentation-on-mevisURVOS#15J&F: 27.8J: 25.7F: 29.9
referring-video-object-segmentation-on-refURVOS#10J&F: 51.6J: 47.3F: 56.0