LAVT: Language-Aware Vision Transformer for Referring Image Segmentation

Benchmark Model Rank Results
generalized-referring-expression-segmentation-on-grefcocoLAVT#12gIoU: 58.40cIoU: 57.64
referring-expression-segmentation-on-refcocog-testLAVT (Swin-B)#14Overall IoU: 62.09
referring-expression-segmentation-on-refcoco-plus-test-bLAVT#17Overall IoU: 55.1
referring-expression-segmentation-on-refcocog-valLAVT#20Overall IoU: 61.24
referring-expression-segmentation-on-refcoco-plus-testaLAVT#21Overall IoU: 68.38
referring-expression-segmentation-on-refcoco-plus-valLAVT#26Overall IoU: 62.14