Scene Recognition on Places-365
60.7Top-1 AccVIC-MAE
Evaluation results
| Method | Links | |
|---|---|---|
| VIC-MAEArch.=ViT-L, Pre-training Data=K710 + MiT + IN1K, Evaluation Protocol=Self-Supervised2023.03 | 60.7 | |
| Absolute SOTAReference=[67]2023.05 | 60.7 | |
| VIC-MAEArch.=ViT-L, Pre-training Data=K400 + IN1K, Evaluation Protocol=Self-Supervised2023.03 | 60 | |
| VIC-MAEArch.=ViT-L, Pre-training Data=MiT, Evaluation Protocol=Self-Supervised2023.03 | 59.7 | |
| VIC-MAEArch.=ViT-L, Pre-training Data=K400, Evaluation Protocol=Self-Supervised2023.03 | 59.5 | |
| MAEArch.=ViT-L, Pre-training Data=IN1K, Evaluation Protocol=Self-Supervised2023.03 | 59.4 | |
| OmniMAEArch.=ViT-L, Pre-training Data=K400 + IN1K, Evaluation Protocol=Self-Supervised2023.03 | 59.4 | |
| OMNIVOREArch.=ViT-B, Pre-training Data=IN1K + K400 + SUN RGB-D, Evaluation Protocol=Supervised2023.03 | 59.2 | |
| VIC-MAEArch.=ViT-B, Pre-training Data=K710 + MiT + IN1K, Evaluation Protocol=Self-Supervised2023.03 | 59.1 | |
| ViTArch.=ViT-L, Pre-training Data=IN1K, Evaluation Protocol=Supervised2023.03 | 58.9 | |
| VIC-MAEArch.=ViT-B, Pre-training Data=K400 + IN1K, Evaluation Protocol=Self-Supervised2023.03 | 58.6 | |
| OmniMAEArch.=ViT-B, Pre-training Data=K400 + IN1K, Evaluation Protocol=Self-Supervised2023.03 | 58.5 | |
| ST-MAEArch.=ViT-L, Pre-training Data=K400, Evaluation Protocol=Self-Supervised2023.03 | 58.1 | |
| MAEArch.=ViT-B, Pre-training Data=IN1K, Evaluation Protocol=Self-Supervised2023.03 | 57.9 | |
| ST-MAEArch.=ViT-B, Pre-training Data=K400, Evaluation Protocol=Self-Supervised2023.03 | 57.4 | |
| ViTArch.=ViT-B, Pre-training Data=IN1K, Evaluation Protocol=Supervised2023.03 | 57 | |
| SigLIP2-B/16-256Evaluation protocol=frozen-backbone linear probing, Crop size=256 x 2562026.05 | 56.18 | |
| VECA-BEvaluation protocol=frozen-backbone linear probing, Crop size=256 x 256, Core budget (C)=642026.05 | 55.82 | |
| DFN2B-CLIP-B/16Evaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.05 | 55.79 | |
| DINOv3-BEvaluation protocol=frozen-backbone linear probing, Crop size=256 x 2562026.05 | 55.39 | |
| OpenCLIP-B/16Evaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.05 | 55.35 | |
| VECA-BEvaluation protocol=frozen-backbone linear probing, Crop size=256 x 256, Core budget (C)=82026.05 | 55.3 | |
| CLIP-B/16Evaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.05 | 55.27 | |
| DINOv2-Reg-BEvaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.05 | 54.81 | |
| C-RADIOv3-BEvaluation protocol=frozen-backbone linear probing, Crop size=256 x 2562026.05 | 54.13 | |
| DINOv2-BEvaluation protocol=frozen-backbone linear probing, Crop size=224 x 2242026.05 | 53.37 | |
| IMAGEBINDZero-shot=true, Text Prompting=true2023.05 | 45.4 | |
| RandomZero-shot=true2023.05 | 0.27 |