Image Classification on ImageNet (val) (Top-1)
83.3Top-1 AccSwin-B
Evaluation Results
| Method | Links | |
|---|---|---|
| Swin-BParams (M)=88, GFLOPs=15.4, Resolution=2242021.11 | 83.3 | |
| LV-VIT-SParams (M)=26.2, GFLOPs=6.6, Resolution=2242021.11 | 83.3 | |
| CvT-21Params (M)=32, GFLOPs=24.9, Resolution=3842021.11 | 83.3 | |
| CvT-21+ATSParams (M)=32, GFLOPs=17.4, Resolution=3842021.11 | 83.1 | |
| Swin-SParams (M)=50, GFLOPs=8.7, Resolution=2242021.11 | 83 | |
| CvT-13Params (M)=20, GFLOPs=16.3, Resolution=3842021.11 | 83 | |
| RegNetY-16GParams (M)=84, GFLOPs=16, Resolution=2242021.11 | 82.9 | |
| CvT-13+ATSParams (M)=20, GFLOPs=11.7, Resolution=3842021.11 | 82.9 | |
| TNT-BParams (M)=66, GFLOPs=14.1, Resolution=2242021.11 | 82.8 | |
| CvT-21Params (M)=32, GFLOPs=7.1, Resolution=2242021.11 | 82.5 | |
| T2T-ViT-24Params (M)=64.1, GFLOPs=14.1, Resolution=2242021.11 | 82.3 | |
| PS-VIT-B/18Params (M)=21.3, GFLOPs=8.8, Resolution=2242021.11 | 82.3 | |
| CvT-21+ATSParams (M)=32, GFLOPs=5.1, Resolution=2242021.11 | 82.3 | |
| CrossViT-BParams (M)=104.7, GFLOPs=21.2, Resolution=2242021.11 | 82.2 | |
| PS-VIT-B/18+ATSParams (M)=21.3, GFLOPs=5.6, Resolution=2242021.11 | 82.2 | |
| CoaT-Lite SmallParams (M)=20, GFLOPs=4, Resolution=2242021.11 | 81.9 | |
| T2T-ViT-19Params (M)=39.2, GFLOPs=8.9, Resolution=2242021.11 | 81.9 | |
| DeiT-Base/16Params (M)=86.6, GFLOPs=17.6, Resolution=2242021.11 | 81.8 | |
| PS-VIT-B/14Params (M)=21.3, GFLOPs=5.4, Resolution=2242021.11 | 81.7 | |
| RegNet Y-8GParams (M)=39, GFLOPs=8, Resolution=2242021.11 | 81.7 | |
| CvT-13Params (M)=20, GFLOPs=4.5, Resolution=2242021.11 | 81.6 | |
| T2T-ViT-14Params (M)=22, GFLOPs=5.2, Resolution=2242021.11 | 81.5 | |
| CPVT-Small-GAPParams (M)=23, GFLOPs=4.6, Resolution=2242021.11 | 81.5 | |
| PS-ViT-B/14+ATSParams (M)=21.3, GFLOPs=3.7, Resolution=2242021.11 | 81.5 | |
| CvT-13+ATSParams (M)=20, GFLOPs=3.2, Resolution=2242021.11 | 81.4 | |
| Swin-TParams (M)=29, GFLOPs=4.5, Resolution=2242021.11 | 81.3 | |
| PVT-MediumParams (M)=44.2, GFLOPs=6.7, Resolution=2242021.11 | 81.2 | |
| CrossViT-SParams (M)=26.7, GFLOPs=5.6, Resolution=2242021.11 | 81 | |
| CoaT MiniParams (M)=10, GFLOPs=6.8, Resolution=2242021.11 | 80.8 | |
| DeiT-SParams (M)=22.05, GFLOPs=4.6, Resolution=2242021.11 | 79.8 | |
| PVT-SmallParams (M)=24.5, GFLOPs=3.8, Resolution=2242021.11 | 79.8 | |
| DeiT-S+ATSParams (M)=22.05, GFLOPs=2.9, Resolution=2242021.11 | 79.7 | |
| EVIT-DeiT-S (30 epochs)Params (M)=22.1, GFLOPs=3, Resolution=2242021.11 | 79.5 | |
| DynamicViT-DeiT-S (30 Epochs)Params (M)=22.77, GFLOPs=2.9, Resolution=2242021.11 | 79.3 | |
| IA-RED²GFLOPs=2.9, Resolution=2242021.11 | 78.6 | |
| HVT-S-1Params (M)=22.09, GFLOPs=2.4, Resolution=2242021.11 | 78 | |
| ViT-Base/16Params (M)=86.6, GFLOPs=17.6, Resolution=2242021.11 | 77.9 |