rPPG estimation on Unified 10-second inference benchmark NVIDIA RTX 4090 GPU
0.77Parameters (M)PhysNet
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| PhysNetHardware=NVIDIA RTX 4090 GPU (24GB), Duration=10-second inference2026.04 | 0.77 | 65.74 | 68.73 | 14.55 | 3,750 | |
| FreqPhysHardware=NVIDIA RTX 4090 GPU (24GB), Duration=10-second inference2026.04 | 0.86 | 7.75 | 85.44 | 11.7 | 874 | |
| EfficientPhysHardware=NVIDIA RTX 4090 GPU (24GB), Duration=10-second inference2026.04 | 1.91 | 56.06 | 41.36 | 24.18 | 7,814 | |
| DeepPhysHardware=NVIDIA RTX 4090 GPU (24GB), Duration=10-second inference2026.04 | 1.98 | 111.67 | 28.89 | 34.61 | 10,638 | |
| TS-CANHardware=NVIDIA RTX 4090 GPU (24GB), Duration=10-second inference2026.04 | 1.98 | 111.67 | 26.23 | 38.13 | 11,834 | |
| RhythmMambaHardware=NVIDIA RTX 4090 GPU (24GB), Duration=10-second inference2026.04 | 2 | 12.41 | 27.16 | 36.82 | 2,450 | |
| PhysDiffHardware=NVIDIA RTX 4090 GPU (24GB), Duration=10-second inference2026.04 | 2.64 | 22.46 | 60.23 | 16.6 | 1,246 | |
| PhysFormerHardware=NVIDIA RTX 4090 GPU (24GB), Duration=10-second inference2026.04 | 7.38 | 47.44 | 50.79 | 19.69 | 6,480 |