List your company
ResearchFormerly WizWand

Multi-label Image Classification on MS-COCO

90.3mAPQ2L-TResL
79.192882.076484.9687.8436Apr 22, 2021Feb 23, 2022Dec 27, 2022Oct 31, 2023Sep 2, 2024Jul 6, 2025May 10, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Q2L-TResLArchitecture=TResNetL(22k), Input Resolution=640x6402021.0790.3
TResNetArchitecture=TResNetL(22k), Input Resolution=640x6402021.0789.8
Q2L-TResXLArchitecture=TResNetXL, Input Resolution=640x6402021.0789
ASLArchitecture=TResNetXL, Input Resolution=640x6402021.0788.4
IN1k-Mul E2EBackbone=ViT-B/16, Resolution=224, Supervision=Multi-label (Mul), Training Protocol=End-to-end2026.0384.7
IN21k-Mul PretrainBackbone=ViT-B/16, Resolution=224, Pre-training=ImageNet-21K (MIIL), Supervision=Multi-label (Mul)2026.0382.8
D2ACEArchitecture=HST2026.0582.52
+IN1k-Mul FTBackbone=ViT-B/16, Resolution=224, Pre-training=ImageNet-21K (MIIL), Supervision=Multi-label (Mul), Training Protocol=Fine-tuning2026.0382.5
+IN1k-Sig FTBackbone=ViT-B/16, Resolution=224, Pre-training=ImageNet-21K (MIIL), Supervision=Single-label (Sig), Training Protocol=Fine-tuning2026.0382.3
DIHCLArchitecture=HST2026.0581.86
Hard-ImbArchitecture=HST2026.0581.82
ML-UncArchitecture=HST2026.0581.79
D2ACEArchitecture=ML-GCN2026.0581.42
ImageNet-21K PretrainPre-training Dataset=ImageNet-21K, Pre-training Method=Semantic Softmax2021.0481.3
RecentArchitecture=HST2026.0581.27
DIHCLArchitecture=ML-GCN2026.0581.03
ML-UncArchitecture=ML-GCN2026.0580.85
RecentArchitecture=ML-GCN2026.0580.74
BalanceArchitecture=HST2026.0580.65
Hard-ImbArchitecture=ML-GCN2026.0580.55
ActiveArchitecture=HST2026.0580.54
Open Images PretrainPre-training Dataset=Open Images (v6), Pre-training Method=Multi-label training2021.0480.5
ActiveArchitecture=ML-GCN2026.0580.25
BalanceArchitecture=ML-GCN2026.0580.21
RandomArchitecture=HST2026.0580.12
RandomArchitecture=ML-GCN2026.0579.62