Pedestrian Detection on Caltech (test)
9.6MRRPN+BF
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| RPN+BFhardware=Tesla K40 GPU, time/img (s)=0.52016.07 | 9.6 | — | — | |
| SA-FasterRCNN2018.10 | 9.7 | — | — | |
| CompACT-Deephardware=Tesla K40 GPU, time/img (s)=0.52016.07 | 11.7 | — | — | |
| CSPTraining Dataset=CityPersons, Evaluation Setting=reasonable2019.04 | 11.9 | — | — | |
| Part-Level CNN2018.10 | 12.4 | — | — | |
| Checkerboards+2018.10 | 17.1 | — | — | |
| CCFhardware=Titan Z GPU, time/img (s)=132016.07 | 17.3 | — | — | |
| ALFNetTraining Dataset=CityPersons, Evaluation Setting=reasonable2019.04 | 17.8 | — | — | |
| TA-CNN2018.10 | 20.9 | — | — | |
| AlexNetFine-tuning=10x, Training proposals=SCF, Testing proposals=Katamari2015.01 | 21.6 | 0.9 | — | |
| SpatialPooling+2015.01 | 21.9 | — | — | |
| AlexNetFine-tuning=10x, Training proposals=SCF, Testing proposals=SP+2015.01 | 22 | -0.1 | — | |
| Katamari2015.01 | 22.5 | — | — | |
| AlexNetFine-tuning=10x, Training proposals=SCF, Testing proposals=SCF2015.01 | 23.3 | 11.5 | — | |
| AlexNet2018.10 | 23.3 | — | — | |
| AlexNetFine-tuning=10x, Training proposals=SCF, Testing proposals=LDCF2015.01 | 23.4 | 1.4 | — | |
| AlexNetFine-tuning=1x, Training proposals=SCF, Testing proposals=Katamari2015.01 | 24.2 | -1.7 | — | |
| LDCF2015.01 | 24.8 | — | — | |
| LDCFhardware=CPU, time/img (s)=0.62016.07 | 24.8 | — | — | |
| LDCF2018.10 | 24.8 | — | — | |
| AlexNetFine-tuning=1x, Training proposals=ACF, Testing proposals=Katamari2015.01 | 25.1 | -2.6 | — | |
| AlexNetFine-tuning=1x, Training proposals=SCF, Testing proposals=SCF2015.01 | 25.9 | 8.9 | — | |
| AlexNetFine-tuning=1x, Training proposals=ACF, Testing proposals=SCF2015.01 | 26.9 | 7.9 | — | |
| AlexNetArchitecture=AlexNet, Number of parameters=~10^7, Training data density=Caltech10x2015.01 | 27.5 | — | — | |
| MediumNetArchitecture=MediumNet, Number of parameters=~10^6, Training data density=Caltech10x2015.01 | 27.9 | — | — | |
| CifarNetArchitecture=CifarNet, Number of parameters=~10^5, Training data density=Caltech10x2015.01 | 28.4 | — | — | |
| CifarNet2018.10 | 28.4 | — | — | |
| CifarNetArchitecture=CifarNet, Number of parameters=~10^5, Training data density=Caltech1x2015.01 | 30.7 | — | — | |
| AlexNetArchitecture=AlexNet, Number of parameters=~10^7, Training data density=Caltech1x2015.01 | 32.4 | — | — | |
| AlexNetFine-tuning=1x, Training proposals=SCF, Testing proposals=ACF2015.01 | 34.3 | 9.9 | — | |
| AlexNetFine-tuning=1x, Training proposals=ACF, Testing proposals=ACF2015.01 | 34.5 | 9.7 | — | |
| SquaresChnFtrs2015.01 | 34.8 | — | — | |
| SquaresChnFtrsArchitecture=SquaresChnFtrs, Training data density=Caltech1x2015.01 | 34.8 | — | — | |
| SDN2018.10 | 37.9 | — | — | |
| JointDeep2018.10 | 39.3 | — | — | |
| ACF2015.01 | 44.2 | — | — | |
| SOTAAdditional Data=true2023.03 | — | — | 28.8 | |
| UniHCPProtocol=Direct Evaluation2023.03 | — | — | 37.8 | |
| UniHCPProtocol=Fine-tuning2023.03 | — | — | 27.2 |