Video Semantic Segmentation on CaDIS
95.29F1 (Hand)TCNN
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| TCNNBackbone=ResNet-50, Additional Supervision=Autoencoder Network2021.12 | 95.29 | 83.58 | 90.06 | 83.22 | 96.55 | 90.77 | 90.22 | 95.21 | 90.61 | 83.18 | 93.3 | |
| Image PredictorBackbone=ResNet-50, Base Model=DeepLabv3+2021.12 | 94.59 | 80.31 | 88.82 | 79.68 | 96.51 | 86.75 | 89.8 | 93.58 | 88.76 | 80.28 | 91.62 | |
| Video PredictorBackbone=ResNet-50, Base Model=DeepLabv3+, Temporal Module=3x3 ConvLSTM2021.12 | 92.73 | 82.4 | 89.52 | 81.72 | 96.26 | 90.81 | 89.98 | 95.17 | 89.82 | 81.9 | 93.11 |