MaskRIS: Semantic Distortion-aware Data Augmentation for Referring Image Segmentation

Open paper
Benchmark Model Rank Results
referring-expression-segmentation-on-refcocoMaskRIS (Swin-B, combined DB)#13Overall IoU: 78.71
referring-expression-segmentation-on-refcocoMaskRIS (Swin-B)#14Overall IoU: 76.49Mean IoU: 78.35
referring-expression-segmentation-on-refcoco-3MaskRIS (Swin-B, combined DB)#12Overall IoU: 70.26
referring-expression-segmentation-on-refcoco-3MaskRIS (Swin-B)#16Overall IoU: 67.54Mean IoU: 71.68
referring-expression-segmentation-on-refcoco-4MaskRIS (Swin-B, combined DB)#10Overall IoU: 75.15
referring-expression-segmentation-on-refcoco-4MaskRIS (Swin-B)#12Overall IoU: 74.46Mean IoU: 76.73
referring-expression-segmentation-on-refcoco-5MaskRIS (Swin-B, combined DB)#9Overall IoU: 62.83
referring-expression-segmentation-on-refcoco-5MaskRIS (Swin-B)#11Overall IoU: 59.39Mean IoU: 64.5
referring-expression-segmentation-on-refcoco-8MaskRIS (Swin-B, combined DB)#7Overall IoU: 80.64
referring-expression-segmentation-on-refcoco-8MaskRIS (Swin-B)#8Overall IoU: 78.96Mean IoU: 80.24
referring-expression-segmentation-on-refcoco-9MaskRIS (Swin-B, combined DB)#7Overall IoU: 75.1
referring-expression-segmentation-on-refcoco-9MaskRIS (Swin-B)#8Overall IoU: 73.96Mean IoU: 76.06
referring-expression-segmentation-on-refcocogMaskRIS (Swin-B, combined DB)#11Overall IoU: 69.12
referring-expression-segmentation-on-refcocogMaskRIS (Swin-B)#13Overall IoU: 65.55Mean IoU: 69.31
referring-expression-segmentation-on-refcocog-1MaskRIS (Swin-B, combined DB)#8Overall IoU: 71.09
referring-expression-segmentation-on-refcocog-1MaskRIS (Swin-B)#11Overall IoU: 66.5Mean IoU: 69.42

Rank counts only results with a code link.