List your company
ResearchFormerly WizWand

Referring Image Segmentation on RefCOCO RefCOCO+ RefCOCOg Overall Average

42.33oIoU+B2Gfix
31.118834.029436.9439.8506Sep 27, 2025
Updated 7mo ago

Evaluation results

MethodLinks
+B2GfixBackbone=DFN VIT-H/14, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0942.33
+B2Gdyn-clusBackbone=DFN VIT-H/14, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0942.31
+B2Gdyn-permBackbone=DFN VIT-H/14, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0942.02
+B2GfixBackbone=CLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0941.85
+B2Gdyn-permBackbone=CLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0941.75
+B2Gdyn-clusBackbone=CLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0941.58
+B2Gdyn-clusBackbone=SigLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0941.54
+B2Gdyn-permBackbone=SigLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0941.24
+B2GfixBackbone=SigLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0941.05
Global-LocalBackbone=SigLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0940.69
+B2GfixBackbone=CLIP ViT-B/32, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0939.78
+B2Gdyn-permBackbone=CLIP ViT-B/32, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0939.54
+B2GBackbone=CLIP ViT-B/16, Pre-trained Segmentor=SAM, Evaluation Protocol=Zero-shot2025.0939.5
+B2Gdyn-clusBackbone=CLIP ViT-B/32, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0939.38
HybridGLBackbone=CLIP ViT-B/16, Pre-trained Segmentor=SAM, Evaluation Protocol=Zero-shot2025.0938.68
Global-LocalBackbone=CLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0938.58
HybridGLBackbone=CLIP ViT-B/16, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0938.02
Global-LocalBackbone=DFN VIT-H/14, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0936.74
HybridGLBackbone=CLIP ViT-B/32, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0936.73
Global-LocalBackbone=CLIP ViT-B/32, Pre-trained Segmentor=Mask2Former, Evaluation Protocol=Zero-shot2025.0936.55
+B2GBackbone=CLIP ViT-B/32, Pre-trained Segmentor=SAM, Evaluation Protocol=Zero-shot2025.0935.23
Global-LocalBackbone=CLIP ViT-B/16, Pre-trained Segmentor=SAM, Evaluation Protocol=Zero-shot2025.0935.19
HybridGLBackbone=CLIP ViT-B/32, Pre-trained Segmentor=SAM, Evaluation Protocol=Zero-shot2025.0934.9
Global-LocalBackbone=CLIP ViT-B/32, Pre-trained Segmentor=SAM, Evaluation Protocol=Zero-shot2025.0931.55