Car Counting on PUCPR+ (test)
2.42MAESAFECount
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SAFECountMethod category=Few-shot counting method2022.01 | 2.42 | 3.55 | |
| HLCNNMethod category=Single-class car counting method2022.01 | 2.52 | 3.4 | |
| Our full approachbackbone=RetinaNet, components=Soft-IoU layer, EM-Merger unit2019.04 | 7.16 | 12 | |
| PDEMMethod category=Multi-class counting method2022.01 | 7.16 | 12 | |
| FamNetMethod category=Few-shot counting method, Evaluation protocol=trained and evaluated by current authors with official code2022.01 | 14.68 | 19.38 | |
| IEP Countingyear=20192019.04 | 15.17 | — | |
| IEP CountMethod category=Multi-class counting method2022.01 | 15.17 | — | |
| One-Look RegressionRevised=true2017.07 | 21.88 | 36.73 | |
| One-Look Regressionyear=20162019.04 | 21.88 | 36.73 | |
| One LookMethod category=Multi-class counting method2022.01 | 21.88 | 36.73 | |
| Our Car Counting CNN ModelNp=4002017.07 | 22.76 | 34.46 | |
| LPN Countingyear=20172019.04 | 22.76 | 34.46 | |
| LPNMethod category=Single-class car counting method2022.01 | 22.76 | 34.46 | |
| RetinaNetyear=20182019.04 | 24.58 | 33.12 | |
| RetinaNetMethod category=Detector provided by the benchmark2022.01 | 24.58 | 33.12 | |
| Faster R-CNN (RPN-small)Np=400, Fine-tuned=true2017.07 | 39.88 | 47.67 | |
| Faster R-CNNyear=20152019.04 | 39.88 | 47.67 | |
| S-RPNMethod category=Detector provided by the benchmark2022.01 | 39.88 | 47.67 | |
| VLCounterEvaluation Protocol=zero-shot, Stage=1, Class Name=Checkmark, Train Dataset=FSC1472023.12 | 48.94 | 69.08 | |
| BMNet+Evaluation Protocol=few-shot, Stage=1, Class Name=X, Train Dataset=FSC1472023.12 | 62.42 | 81.74 | |
| FamNetEvaluation Protocol=few-shot, Stage=1, Class Name=X, Train Dataset=FSC1472023.12 | 87.54 | 117.68 | |
| VLBaseEvaluation Protocol=zero-shot, Stage=1, Class Name=Checkmark, Train Dataset=FSC1472023.12 | 90.82 | 104.01 | |
| BMNetEvaluation Protocol=few-shot, Stage=1, Class Name=X, Train Dataset=FSC1472023.12 | 103.18 | 112.42 | |
| Faster R-CNNNp=400, Fine-tuned=true2017.07 | 111.4 | 149.35 | |
| F-RCNNMethod category=Detector provided by the benchmark2022.01 | 111.4 | 149.35 | |
| YOLO9000optyear=20172019.04 | 130.4 | 172.46 | |
| YOLOFine-tuned=true2017.07 | 156 | 200.42 | |
| YOLOyear=20162019.04 | 156 | 200.42 | |
| YOLOMethod category=Detector provided by the benchmark2022.01 | 156 | 200.42 | |
| YOLOFine-tuned=false2017.07 | 156.72 | 200.54 | |
| Faster R-CNNNp=400, Fine-tuned=false2017.07 | 156.76 | 200.59 |