Video Alignment on Penn Action
0.9728Kendall's TauCASA
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CASAPose=true, ImageNet pre-trained=false2022.04 | 0.9728 | 0.9449 | |
| VAVA2021.11 | 0.8053 | 0.7091 | |
| LAV2021.11 | 0.8047 | 0.6613 | |
| LAVPose=false, ImageNet pre-trained=true2022.04 | 0.8047 | 0.6613 | |
| Representation Learning via Global Temporal Alignment and Cycle-Consistencyper-class training=true2021.05 | 0.7829 | — | |
| GTA2021.11 | 0.7829 | — | |
| HadjiPose=false, ImageNet pre-trained=false2022.04 | 0.7829 | — | |
| TCC + TCN2019.12 | 0.7672 | — | |
| Pr-VIPEevaluation_subset=different-view only2019.12 | 0.7607 | — | |
| Pr-VIPEevaluation_subset=same-view only2019.12 | 0.7521 | — | |
| Ourstraining=joint representation2021.05 | 0.7484 | — | |
| Pr-VIPE2019.12 | 0.7476 | — | |
| Pr-VIPEPose=2D poses, ImageNet pre-trained=false2022.04 | 0.7476 | — | |
| TCN2019.12 | 0.7353 | — | |
| TCCper-class training=true2021.05 | 0.7353 | — | |
| TCC2021.11 | 0.7353 | 0.6726 | |
| TCC2019.12 | 0.7328 | — | |
| TCNper-class training=true2021.05 | 0.7328 | — | |
| TCN2021.11 | 0.7328 | 0.6762 | |
| TCNPose=false, ImageNet pre-trained=true2022.04 | 0.7328 | 0.6762 | |
| TCC + SaL2019.12 | 0.7286 | — | |
| TCCtraining=joint representation2021.05 | 0.71 | — | |
| TCCPose=false, ImageNet pre-trained=true2022.04 | 0.7012 | 0.6638 | |
| LAVPose=true, ImageNet pre-trained=true2022.04 | 0.6983 | 0.6404 | |
| TCNtraining=joint representation2021.05 | 0.6529 | — | |
| SaL2019.12 | 0.6336 | — | |
| SaLper-class training=true2021.05 | 0.6336 | — | |
| SAL2021.11 | 0.6336 | 0.5943 | |
| SaLPose=false, ImageNet pre-trained=true2022.04 | 0.6336 | 0.5943 | |
| TCCPose=true, ImageNet pre-trained=true2022.04 | 0.6267 | 0.6268 | |
| SaLtraining=joint representation2021.05 | 0.5387 | — | |
| LAVPose=true, ImageNet pre-trained=false2022.04 | 0.4929 | 0.3853 | |
| TCCPose=true, ImageNet pre-trained=false2022.04 | 0.4529 | 0.4304 |