VLT: Vision-Language Transformer and Query Generation for Referring Segmentation

Benchmark Model Rank Results
referring-video-object-segmentation-on-refer-youtube-vosVLT#10J&F: 63.8J: 61.9F: 65.6
referring-expression-segmentation-on-refcoco-plus-test-bVLT#16Overall IoU: 56.92
referring-expression-segmentation-on-refcocog-valVLT (Swin-B)#18Overall IoU: 63.49
referring-expression-segmentation-on-refcoco-plus-testaVLT#20Overall IoU: 68.43
referring-expression-segmentation-on-refcoco-plus-valVLT#24Overall IoU: 63.53
referring-expression-segmentation-on-refcoco-valVLT#24Overall IoU: 72.96
referring-expression-segmentation-on-refer-youtube-vos-2021-public-validationVLT#32J&F: 63.8J: 61.9F: 65.6
referring-video-object-segmentation-on-mevisVLT+TC#32J&F: 35.5J: 33.6F: 37.3