Bird S Eye View Semantic Segmentation On Lyft
Metrics
IoU vehicle - 224x480 - Long
IoU vehicle - 224x480 - Short
Results
Performance results of various models on this benchmark
Model Name | IoU vehicle - 224x480 - Long | IoU vehicle - 224x480 - Short | Paper Title | Repository |
|---|---|---|---|---|
| PointBeV (ResNet-50) | 44.5 | 72.3 | PointBeV: A Sparse Approach to BeV Predictions | |
| BEVFormer (EfficientNet-b4) | 44.5 | 69.9 | BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers | |
| Simple-BEV (EfficientNet-b4) | 44.5 | 70.4 | Simple-BEV: What Really Matters for Multi-Sensor BEV Perception? | |
| PointBeV (EfficientNet-b4) | 45.4 | 72.6 | PointBeV: A Sparse Approach to BeV Predictions | |
| FIERY | 36.7 | 59.4 | FIERY: Future Instance Prediction in Bird's-Eye View from Surround Monocular Cameras | |
| Simple-BEV (ResNet-50) | 43.6 | 70.7 | Simple-BEV: What Really Matters for Multi-Sensor BEV Perception? | |
| BEVFormer(ResNet-50) | 43.2 | 68.8 | BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers |
0 of 7 row(s) selected.