Driver and Traffic Scene Recognition on AIDE
77.39DER AccuracyUV-M3TL
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| UV-M3TLBackbone (Face)=MARNet, Backbone (Body)=MARNet, Backbone (Gesture)=3DCNN, Backbone (Posture)=3DCNN, Backbone (Scene)=MARNet2026.02 | 77.39 | 73.82 | 96.57 | 87.07 | 83.71 | |
| MMTL-UniADBackbone (Face)=MARNet, Backbone (Body)=MARNet, Backbone (Gesture)=3DCNN, Backbone (Posture)=3DCNN, Backbone (Scene)=MARNet2026.02 | 76.67 | 73.61 | 93.91 | 85 | 82.3 | |
| PMANetPattern=2D, Backbone (Face)=PMANet, Backbone (Body)=PMANet, Backbone (Gesture)=3DCNN, Backbone (Posture)=3DCNN, Backbone (Scene)=PMANet2026.02 | 76.56 | 71.88 | 95.31 | 82.81 | 81.64 | |
| HATPattern=2D, Backbone (Face)=HAT, Backbone (Body)=HAT, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=HAT2026.02 | 75 | 70.31 | 92.18 | 81.25 | 79.69 | |
| MTS-MambaPattern=2D, Backbone (Face)=MTS-Mamba, Backbone (Body)=MTS-Mamba, Backbone (Gesture)=3DCNN, Backbone (Posture)=3DCNN, Backbone (Scene)=MTS-Mamba2026.02 | 75 | 69.31 | 96.29 | 86.11 | 81.68 | |
| TimeSFormerPattern=3D, Backbone (Face)=TimeSFormer, Backbone (Body)=TimeSFormer, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=TimeSFormer2026.02 | 74.87 | 65.18 | 92.12 | 78.81 | 77.75 | |
| YOLOPPattern=2D, Backbone (Face)=YOLOP, Backbone (Body)=YOLOP, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=YOLOP2026.02 | 73.56 | 69.18 | 91.38 | 79.26 | 78.35 | |
| Video Swin TransformerPattern=3D, Backbone (Face)=Video Swin Transformer, Backbone (Body)=Video Swin Transformer, Backbone (Gesture)=3DCNN, Backbone (Posture)=3DCNN, Backbone (Scene)=Video Swin Transformer2026.02 | 73.44 | 65.63 | 93.75 | 75 | 76.96 | |
| Res34+TransEPattern=2D + Sequential, Backbone (Face)=Res18+TransE, Backbone (Body)=Res34+TransE, Backbone (Gesture)=MLP+TE, Backbone (Posture)=MLP+TE, Backbone (Scene)=Res34+TransE2026.02 | 72.65 | 67.08 | 86.63 | 78.46 | 76.21 | |
| ShuffleNet-V1Pattern=3D, Backbone (Face)=ShuffleNet-V1, Backbone (Body)=ShuffleNet-V1, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=ShuffleNet-V12026.02 | 72.41 | 68.97 | 90.64 | 80.79 | 78.2 | |
| SlowFastPattern=3D, Backbone (Face)=SlowFast, Backbone (Body)=SlowFast, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=SlowFast2026.02 | 72.38 | 61.58 | 86.86 | 78.33 | 74.79 | |
| MobileNet-V1Pattern=3D, Backbone (Face)=MobileNet-V1, Backbone (Body)=MobileNet-V1, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=MobileNet-V12026.02 | 72.23 | 64.2 | 88.34 | 77.83 | 75.65 | |
| GLMDriveNetPattern=2D, Backbone (Face)=GLMDriveNet, Backbone (Body)=GLMDriveNet, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=GLMDriveNet2026.02 | 71.38 | 66.57 | 90.23 | 77.19 | 76.34 | |
| Res34 BaselinePattern=2D, Backbone (Face)=Res18, Backbone (Body)=Res34, Backbone (Gesture)=MLP+SE, Backbone (Posture)=MLP+SE, Backbone (Scene)=Res342026.02 | 71.26 | 65.35 | 83.74 | 77.12 | 74.37 | |
| VGG16+TransEPattern=2D + Sequential, Backbone (Face)=VGG13+TransE, Backbone (Body)=VGG16+TransE, Backbone (Gesture)=MLP+TE, Backbone (Posture)=MLP+TE, Backbone (Scene)=VGG16+TransE2026.02 | 71.12 | 67.15 | 85.13 | 78.58 | 75.5 | |
| ShuffleNet-V2Pattern=3D, Backbone (Face)=ShuffleNet-V2, Backbone (Body)=ShuffleNet-V2, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=ShuffleNet-V22026.02 | 70.94 | 64.04 | 89.33 | 78.98 | 75.82 | |
| I3DPattern=3D, Backbone (Face)=I3D, Backbone (Body)=I3D, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=I3D2026.02 | 70.94 | 66.17 | 87.68 | 79.81 | 76.15 | |
| PP-Res18+TransEPattern=2D + Sequential, Backbone (Face)=Res18+TransE, Backbone (Body)=Res34+TransE, Backbone (Gesture)=MLP+TE, Backbone (Posture)=MLP+TE, Backbone (Scene)=PP-Res18+TransE2026.02 | 70.83 | 67.32 | 90.54 | 79.97 | 77.17 | |
| VGG16 BaselinePattern=2D, Backbone (Face)=VGG13, Backbone (Body)=VGG16, Backbone (Gesture)=MLP+SE, Backbone (Posture)=MLP+SE, Backbone (Scene)=VGG162026.02 | 70.72 | 63.65 | 82.77 | 77.94 | 73.77 | |
| Res50+TransEPattern=2D + Sequential, Backbone (Face)=Res34+TransE, Backbone (Body)=Res50+TransE, Backbone (Gesture)=MLP+TE, Backbone (Posture)=MLP+TE, Backbone (Scene)=Res50+TransE2026.02 | 70.24 | 65.65 | 82.57 | 77.29 | 73.94 | |
| 3D-Res34Pattern=3D, Backbone (Face)=3D-Res18, Backbone (Body)=3D-Res34, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=3D-Res342026.02 | 70.11 | 66.52 | 88.51 | 81.21 | 76.59 | |
| Res50 BaselinePattern=2D, Backbone (Face)=Res34, Backbone (Body)=Res50, Backbone (Gesture)=MLP+SE, Backbone (Posture)=MLP+SE, Backbone (Scene)=Res502026.02 | 69.68 | 59.77 | 80.13 | 71.26 | 70.21 | |
| VGG19+TransEPattern=2D + Sequential, Backbone (Face)=VGG16+TransE, Backbone (Body)=VGG19+TransE, Backbone (Gesture)=MLP+TE, Backbone (Posture)=MLP+TE, Backbone (Scene)=VGG19+TransE2026.02 | 69.46 | 65.48 | 85.74 | 77.91 | 74.65 | |
| VGG19 BaselinePattern=2D, Backbone (Face)=VGG16, Backbone (Body)=VGG19, Backbone (Gesture)=MLP+SE, Backbone (Posture)=MLP+SE, Backbone (Scene)=VGG192026.02 | 69.31 | 62.34 | 83.58 | 75.13 | 72.59 | |
| AbSViTPattern=2D, Backbone (Face)=AbSViT, Backbone (Body)=AbSViT, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=AbSViT2026.02 | 69.15 | 67.84 | 92.07 | 80.82 | 77.47 | |
| 3D-Res50Pattern=3D, Backbone (Face)=3D-Res34, Backbone (Body)=3D-Res50, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=3D-Res502026.02 | 69.13 | 63.05 | 87.82 | 79.31 | 74.83 | |
| PP-Res18 BaselinePattern=2D, Backbone (Face)=Res18, Backbone (Body)=Res34, Backbone (Gesture)=MLP+SE, Backbone (Posture)=MLP+SE, Backbone (Scene)=PP-Res182026.02 | 69.05 | 63.87 | 88.01 | 78.16 | 74.77 | |
| CPVTPattern=2D, Backbone (Face)=CPVT, Backbone (Body)=CPVT, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=CPVT2026.02 | 69.01 | 67.35 | 91.44 | 79.57 | 76.84 | |
| CMTPattern=2D, Backbone (Face)=CMT, Backbone (Body)=CMT, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=CMT2026.02 | 68.75 | 68.75 | 93.75 | 81.38 | 78.16 | |
| MobileNet-V2Pattern=3D, Backbone (Face)=MobileNet-V2, Backbone (Body)=MobileNet-V2, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=MobileNet-V22026.02 | 68.47 | 61.74 | 86.54 | 78.66 | 73.85 | |
| MFE-SSNetPattern=2D, Backbone (Face)=MFE-SSNet, Backbone (Body)=MFE-SSNet, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=MFE-SSNet2026.02 | 68.12 | 69.3 | 92.33 | 77.32 | 76.77 | |
| GroupMixFormerPattern=2D, Backbone (Face)=GroupMixFormer, Backbone (Body)=GroupMixFormer, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=GroupMixFormer2026.02 | 66.29 | 67.54 | 92.12 | 77.63 | 75.9 | |
| C3DPattern=3D, Backbone (Face)=C3D, Backbone (Body)=C3D, Backbone (Gesture)=ST-GCN, Backbone (Posture)=ST-GCN, Backbone (Scene)=C3D2026.02 | 63.05 | 63.95 | 85.41 | 77.01 | 72.36 |