Image Recognition on ImageNet1K 1.0 (val)
85.5Top-1 AccWave-ViT-L
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Wave-ViT-LParams=57.5M, GFLOPs=14.8, Category=Large, Token Labeling & conv-stem=true2022.07 | 85.5 | 97.3 | |
| VOLO-D3Params=86.3M, GFLOPs=20.6, Category=Large, Token Labeling & conv-stem=true2022.07 | 85.4 | — | |
| VOLO-D2Params=58.7M, GFLOPs=14.1, Category=Large, Token Labeling & conv-stem=true2022.07 | 85.2 | — | |
| Wave-ViT-BParams=33.5M, GFLOPs=7.2, Category=Base, Token Labeling & conv-stem=true2022.07 | 84.8 | 97.1 | |
| VOLO-D1Params=26.6M, GFLOPs=6.8, Category=Base, Token Labeling & conv-stem=true2022.07 | 84.2 | — | |
| LV-ViT-MParams=55.8M, GFLOPs=16, Category=Large, Token Labeling & conv-stem=true2022.07 | 84.1 | 96.7 | |
| SE-CoTNetD-152Params=55.8M, GFLOPs=17, Category=Large, Token Labeling & conv-stem=false2022.07 | 84 | 97 | |
| Wave-ViT-SParams=22.7M, GFLOPs=4.7, Category=Small, Token Labeling & conv-stem=true2022.07 | 83.9 | 96.6 | |
| RegionViT-B+Params=73.8M, GFLOPs=13.6, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.8 | — | |
| Focal-BaseParams=89.8M, GFLOPs=16, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.8 | 96.5 | |
| PVTv2-B5Params=82.0M, GFLOPs=11.8, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.8 | 96.6 | |
| Twins-SVT-LParams=99.3M, GFLOPs=15.1, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.7 | 96.5 | |
| PVTv2-B4Params=62.6M, GFLOPs=10.1, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.6 | 96.7 | |
| CrossViT-15-384Params=28.5M, GFLOPs=21.4, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.5 | — | |
| BoTNet-S1-128Params=75.1M, GFLOPs=19.3, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.5 | 96.5 | |
| Swin-BParams=88.0M, GFLOPs=15.4, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.5 | 96.5 | |
| RegionViT-M+Params=42.0M, GFLOPs=7.9, Category=Base, Token Labeling & conv-stem=false2022.07 | 83.4 | — | |
| CaiT-S36Params=68.4M, GFLOPs=13.9, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.3 | — | |
| Swin-SParams=50.0M, GFLOPs=8.7, Category=Base, Token Labeling & conv-stem=false2022.07 | 83.2 | 96.2 | |
| Twins-SVT-BParams=56.1M, GFLOPs=8.6, Category=Base, Token Labeling & conv-stem=false2022.07 | 83.2 | 96.3 | |
| SE-CoTNetD-101Params=40.9M, GFLOPs=8.5, Category=Base, Token Labeling & conv-stem=false2022.07 | 83.2 | 96.5 | |
| PVTv2-B3Params=45.2M, GFLOPs=6.9, Category=Base, Token Labeling & conv-stem=false2022.07 | 83.2 | 96.5 | |
| RegionViT-BParams=72.7M, GFLOPs=13, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.2 | 96.1 | |
| DeepViT-LParams=58.9M, GFLOPs=12.8, Category=Large, Token Labeling & conv-stem=false2022.07 | 83.1 | — | |
| TNT-BParams=65.6M, GFLOPs=14.1, Category=Large, Token Labeling & conv-stem=false2022.07 | 82.9 | 96.3 | |
| Wave-ViT-SParams=19.8M, GFLOPs=4.3, Category=Small, Token Labeling & conv-stem=false2022.07 | 82.7 | 96.2 | |
| T2T-ViT-24Params=64.1M, GFLOPs=15, Category=Large, Token Labeling & conv-stem=false2022.07 | 82.6 | 95.9 | |
| CvT-21Params=32.0M, GFLOPs=7.1, Category=Base, Token Labeling & conv-stem=false2022.07 | 82.5 | — | |
| ConViT-BParams=86.5M, GFLOPs=16.8, Category=Large, Token Labeling & conv-stem=false2022.07 | 82.4 | 95.9 | |
| ResNeSt-101Params=48.3M, GFLOPs=10.2, Category=Large, Token Labeling & conv-stem=false2022.07 | 82.3 | — | |
| SE-ResNet-152Params=66.8M, GFLOPs=11.6, Category=Large, Token Labeling & conv-stem=false2022.07 | 82.2 | 95.9 | |
| PVTv2-B2Params=25.4M, GFLOPs=4, Category=Small, Token Labeling & conv-stem=false2022.07 | 82 | 96 | |
| CoaT-Lite SmallParams=20.0M, GFLOPs=4, Category=Small, Token Labeling & conv-stem=false2022.07 | 81.9 | 95.5 | |
| T2T-ViT-19Params=39.2M, GFLOPs=8.5, Category=Base, Token Labeling & conv-stem=false2022.07 | 81.9 | 95.7 | |
| DeiT-BParams=86.6M, GFLOPs=17.6, Category=Large, Token Labeling & conv-stem=false2022.07 | 81.8 | 95.6 | |
| Twins-SVT-SParams=24.1M, GFLOPs=2.9, Category=Small, Token Labeling & conv-stem=false2022.07 | 81.7 | 95.6 | |
| BoTNet-S1-59Params=33.5M, GFLOPs=7.3, Category=Base, Token Labeling & conv-stem=false2022.07 | 81.7 | 95.8 | |
| SE-CoTNetD-50Params=23.1M, GFLOPs=4.1, Category=Small, Token Labeling & conv-stem=false2022.07 | 81.6 | 95.8 | |
| T2T-ViT-14Params=21.5M, GFLOPs=4.8, Category=Small, Token Labeling & conv-stem=false2022.07 | 81.5 | 95.7 | |
| RegionViT-Ti+Params=14.3M, GFLOPs=2.7, Category=Small, Token Labeling & conv-stem=false2022.07 | 81.5 | — | |
| ResNeXt101Params=83.5M, GFLOPs=15.6, Category=Large, Token Labeling & conv-stem=false2022.07 | 81.5 | — | |
| ConViT-SParams=27.8M, GFLOPs=5.4, Category=Small, Token Labeling & conv-stem=false2022.07 | 81.3 | 95.7 | |
| ResNet-152Params=60.2M, GFLOPs=11.6, Category=Large, Token Labeling & conv-stem=false2022.07 | 81.3 | 95.5 | |
| Swin-TParams=29.0M, GFLOPs=4.5, Category=Small, Token Labeling & conv-stem=false2022.07 | 81.2 | 95.5 | |
| BoTNet-S1-50Params=20.8M, GFLOPs=4.3, Category=Small, Token Labeling & conv-stem=false2022.07 | 80.4 | 95 | |
| ResNet-101Params=44.6M, GFLOPs=7.9, Category=Base, Token Labeling & conv-stem=false2022.07 | 80 | 95 | |
| ResNet-50Params=25.5M, GFLOPs=4.1, Category=Small, Token Labeling & conv-stem=false2022.07 | 78.3 | 94.3 |