List your company
ResearchFormerly WizWand

Semantic Segmentation on LVIS 92^i

47.2mIoUINSID3
7.6817.9428.238.46Oct 3, 2024Jan 1, 2025Apr 2, 2025Jul 1, 2025Sep 30, 2025Dec 29, 2025Mar 30, 2026
Updated 7mo ago

Evaluation results

MethodLinks
INSID3Encoder=DINOv3, #Param=304 M, Training Protocol=Training free: Unsupervised pre-training, Shot count=52026.0347.2
SegICEncoder=DINOv2, #Param=310 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0344.6
GF-SAMEncoder=DINOv2 + SAM, #Param=945 M, Training Protocol=Training free: Mask-supervised pre-training, Shot count=52026.0344.2
GF-SAM† + our debiasEncoder=DINOv3 + SAM, #Param=945 M, Training Protocol=Training free: Mask-supervised pre-training, Shot count=52026.0343.6
GF-SAM†Encoder=DINOv3 + SAM, #Param=945 M, Training Protocol=Training free: Mask-supervised pre-training, Shot count=52026.0342.8
INSID3Encoder=DINOv3, #Param=304 M, Training Protocol=Training free, Supervision Type=Unsupervised pre-training2026.0341.8
MatcherShots=few-shot, Setting=In-context2024.1040
MatcherEncoder=DINOv2 + SAM, #Param=945 M, Training Protocol=Training free: Mask-supervised pre-training, Shot count=52026.0340
SegIC (COCO)Encoder=DINOv2, #Param=310 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision, dataset split=COCO2026.0335.7
SINEEncoder=DINOv2, #Param=373 M, Training Protocol=Task-specific fine-tuning: Semantic + mask supervision, Shot count=52026.0335.5
DiffewSShots=few-shot, Setting=In-context2024.1035.4
DiffewSEncoder=Stable Diffusion, #Param=890 M, Training Protocol=Task-specific fine-tuning: Semantic + mask supervision, Shot count=52026.0335.4
GF-SAMEncoder=DINOv2 + SAM, #Param=945 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training2026.0335.2
GF-SAM† + debiasEncoder=DINOv3 + SAM, #Param=945 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training, Debias=True2026.0334.6
MatcherShots=one-shot, Setting=In-context2024.1033
MatcherEncoder=DINOv2 + SAM, #Param=945 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training2026.0333
GF-SAM†Encoder=DINOv3 + SAM, #Param=945 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training2026.0331.8
DiffewSShots=one-shot, Setting=In-context2024.1031.4
DiffewSEncoder=Stable Diffusion, #Param=890 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0331.4
SINEEncoder=DINOv2, #Param=373 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0331.2
SegGPTShots=few-shot, Setting=In-context2024.1025.4
SegGPTEncoder=ViT, #Param=354 M, Training Protocol=Task-specific fine-tuning: Semantic + mask supervision, Shot count=52026.0325.4
HSNetShots=few-shot, Setting=In-context2024.1022.9
VATShots=few-shot, Setting=In-context2024.1022.7
SegGPTShots=one-shot, Setting=In-context2024.1018.6
SegGPTEncoder=ViT, #Param=354 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0318.6
VATShots=one-shot, Setting=In-context2024.1018.5
PerSAM-FShots=one-shot, Setting=In-context2024.1018.4
HSNetShots=one-shot, Setting=In-context2024.1017.4
PerSAMShots=one-shot, Setting=In-context2024.1015.6
PerSAM-Fone-shot=true2025.1212.3
PerSAMone-shot=true2025.1211.5
PerSAMEncoder=SAM, #Param=640 M, Training Protocol=Training free, Supervision Type=Mask-supervised pre-training2026.0311.5
PainterShots=few-shot, Setting=In-context2024.1010.9
PainterShots=one-shot, Setting=In-context2024.1010.5
Painterone-shot=true2025.1210.5
PainterEncoder=ViT, #Param=354 M, Training Protocol=Task-specific fine-tuning, Supervision Type=Semantic + mask supervision2026.0310.5
OmniSegNetone-shot=true2025.129.2