List your company
ResearchFormerly WizWand

Part Segmentation on PACO-Part

38.7mIoUINSID3
12.49219.29626.132.904Oct 9, 2024Jan 6, 2025Apr 6, 2025Jul 4, 2025Oct 2, 2025Dec 30, 2025Mar 30, 2026
Updated 7mo ago

Evaluation results

MethodLinks
INSID3Encoder=DINOv3, #Param=304 M, Training Protocol=Training free, Supervision Type=Unsupervised pre-training2026.0338.7
GF-SAMModel Category=generalist model, Shots=1-shot2024.1036.3
GF-SAMEncoder=DINOv2 + SAM, #Param=945 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training2026.0336.3
GF-SAM† + debiasEncoder=DINOv3 + SAM, #Param=945 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training, Debias=True2026.0336.1
MatcherModel Category=generalist model, Shots=1-shot2024.1034.7
MatcherEncoder=DINOv2 + SAM, #Param=945 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training2026.0334.7
GF-SAM†Encoder=DINOv3 + SAM, #Param=945 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training2026.0334.4
SegICEncoder=DINOv2, #Param=310 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0325.9
SegIC (COCO)Encoder=DINOv2, #Param=310 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision, dataset split=COCO2026.0325.1
SINEEncoder=DINOv2, #Param=373 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0323.3
DiffewSEncoder=Stable Diffusion, #Param=890 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0322.8
PerSAM-FModel Category=generalist model, Shots=1-shot2024.1022.7
HSNetModel Category=specialist model, Shots=1-shot2024.1022.6
PerSAMModel Category=generalist model, Shots=1-shot2024.1022.5
PerSAMEncoder=SAM, #Param=640 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training2026.0322.5
PainterEncoder=ViT, #Param=354 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0314.1
SegGPTEncoder=ViT, #Param=354 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0313.5