Lane Detection On Bdd100K Val
평가 지표
Accuracy (%)
IoU (%)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Accuracy (%) | IoU (%) | Paper Title | Repository |
|---|---|---|---|---|
| Enet-SAD | 36.6 | 16.02 | Learning Lightweight Lane Detection CNNs by Self Attention Distillation | |
| YOLOPv2 | 87.8 | 27.25 | YOLOPv2: Better, Faster, Stronger for Panoptic Driving Perception | |
| TwinLiteNetPlus-Small | 75.8 | 29.3 | TwinLiteNetPlus: A Stronger Model for Real-time Drivable Area and Lane Segmentation | |
| TwinLiteNetPlus-Nano | 70.2 | 23.3 | TwinLiteNetPlus: A Stronger Model for Real-time Drivable Area and Lane Segmentation | |
| TwinLiteNet | 77.8 | 31.08 | TwinLiteNet: An Efficient and Lightweight Model for Driveable Area and Lane Segmentation in Self-Driving Cars | |
| A-YOLOM(s) | 84.9 | 28.8 | You Only Look at Once for Real-time and Generic Multi-Task | |
| TwinLiteNetPlus-Large | 81.9 | 34.2 | TwinLiteNetPlus: A Stronger Model for Real-time Drivable Area and Lane Segmentation | |
| TwinLiteNetPlus-Medium | 79.1 | 32.3 | TwinLiteNetPlus: A Stronger Model for Real-time Drivable Area and Lane Segmentation | |
| YOLOP | 70.5 | 26.2 | YOLOP: You Only Look Once for Panoptic Driving Perception | |
| HybridNets | 85.4 | 31.6 | HybridNets: End-to-End Perception Network |
0 of 10 row(s) selected.