TTC forecasting on DAD
0.71MSECollideNet
Evaluation results
| Method | Links | |
|---|---|---|
| CollideNetBackbone=Multi-scale ViT2026.04 | 0.71 | |
| CollideNetTraining=DoTA2026.04 | 0.727 | |
| Video-FocalNetBackbone=Focal Mod. Net2026.04 | 0.73 | |
| CollideNetTraining=CCD2026.04 | 0.731 | |
| VGG-16Backbone=VGG-162026.04 | 0.74 | |
| CNN-RNNBackbone=EfficientNet2026.04 | 0.75 | |
| Video-FocalNetTraining=DoTA2026.04 | 0.76 | |
| VidNeXtBackbone=ConvNeXt2026.04 | 0.77 | |
| Video-FocalNetTraining=CCD2026.04 | 0.787 | |
| TimeSformerBackbone=Vision Transformer2026.04 | 0.79 | |
| Video Swin Trans.Backbone=Swin Transformer2026.04 | 0.79 | |
| VidNeXtTraining=DoTA2026.04 | 0.813 | |
| VidNeXtTraining=CCD2026.04 | 0.817 | |
| X3DBackbone=3D ConvNet2026.04 | 0.85 | |
| ResNet50 3DBackbone=3D ConvNet2026.04 | 0.88 | |
| Li3DBackbone=3D ConvNet2026.04 | 0.9 | |
| HyCTBackbone=CNN-Transformer2026.04 | 0.91 | |
| ViViTBackbone=Vision Transformer2026.04 | 1.02 | |
| C3DBackbone=3D ConvNet2026.04 | 1.75 |