Facial Landmark Detection on 300-W public (test)
2.62NMEGV(HRNet)
Evaluation Results
| Method | Links | |
|---|---|---|
| GV(HRNet)backbone=HRNet2021.11 | 2.62 | |
| GEAN2021.11 | 2.68 | |
| 3DDE2021.11 | 2.69 | |
| LRefNets2021.11 | 2.71 | |
| AWing2021.11 | 2.72 | |
| DCFE2021.11 | 2.76 | |
| HRNetV2021.11 | 2.87 | |
| LAB2021.11 | 2.98 | |
| PFLD 1Xscale=1X2021.11 | 3.01 | |
| ResNet50-FFLDbackbone=ResNet-502021.11 | 3.06 | |
| DAN2021.11 | 3.19 | |
| CCNN2021.11 | 3.23 | |
| SAN2021.11 | 3.34 | |
| efnbackbone=EfficientNet-B32021.11 | 3.34 | |
| CPM2021.11 | 3.39 | |
| ODN2021.11 | 3.56 | |
| mnv2KDbackbone=MobileNetV22021.11 | 3.56 | |
| PCD-CNN2021.11 | 3.67 | |
| mnv2backbone=MobileNetV22021.11 | 3.93 | |
| DSRN2021.11 | 4.12 | |
| RCN2021.11 | 4.67 | |
| RARDL=true, number of parameters=15.65M+, model memory=62.6MB+2018.03 | 4.94 | |
| SIRDL=true, number of parameters=3.72M, model memory=15.6MB2018.03 | 5.04 | |
| MDMDL=true, number of parameters=80.56M, model memory=322.3MB2018.03 | 5.61 | |
| CFSSDL=false, model memory=225.2MB2018.03 | 5.76 | |
| baseline CRDL=true, number of parameters=14.88M, model memory=62.4MB2018.03 | 6.23 | |
| LBFDL=false, model memory=36.6MB2018.03 | 6.32 | |
| RCPRDL=false, model memory=91.3MB2018.03 | 8.35 |