Video Classification on UCF101
93.7AccuracyRSPNet
Evaluation Results
| Method | Links | |
|---|---|---|
| RSPNetBackbone=S3D-G, Extra data=K4002022.11 | 93.7 | |
| EVERESTBackbone=ViT-B, Extra data=None2022.11 | 93.4 | |
| ρBYOLρ=2Backbone=SR50, Extra data=K4002022.11 | 92.7 | |
| VideoMAEBackbone=ViT-B, Extra data=None2022.11 | 91.3 | |
| ρMoCoρ=2Backbone=SR50, Extra data=K4002022.11 | 91 | |
| Vi² CLRBackbone=S3D, Extra data=K4002022.11 | 89.1 | |
| CoCLRBackbone=S3D-G, Extra data=K4002022.11 | 87.9 | |
| ρSwAVρ=2Backbone=SR50, Extra data=K4002022.11 | 87.3 | |
| Vi² CLRBackbone=S3D, Extra data=UCF1012022.11 | 82.8 | |
| CoCLRBackbone=S3D-G, Extra data=UCF1012022.11 | 81.4 | |
| xCLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.10 | 81.1 | |
| nCLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.10 | 79.9 | |
| CLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.10 | 78.9 | |
| CLIP (reported)Pre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.07 | 76.2 | |
| CLIPPre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.07 | 76.1 | |
| MLCDPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.07 | 75.7 | |
| FLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.07 | 74.1 | |
| VCOPBackbone=R(2+1)D, Extra data=UCF1012022.11 | 72.4 | |
| OpenCLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Zero-shot classification2024.07 | 71.3 | |
| I3D_RGB+FlowBackbone=3D Inception2021.03 | 0.98 | |
| BQNBackbone=TSM R502021.03 | 0.976 | |
| TEABackbone=R502021.03 | 0.969 | |
| PAN_FullBackbone=TSM R502021.03 | 0.965 | |
| STMBackbone=R502021.03 | 0.962 | |
| TSMBackbone=R502021.03 | 0.959 | |
| DI Four-StreamBackbone=ResNeXt1012021.03 | 0.955 | |
| TVNetBackbone=BNInception2021.03 | 0.945 | |
| TSN_RGB+FlowBackbone=BNInception2021.03 | 0.94 | |
| StNetBackbone=R502021.03 | 0.935 |