Table 5 Benchmark experiment of MCV-Intention on the state-of-the-art methods.
From: MCV-Intention: A Multimodalities and Cross-View Dataset for Human Assembly Intention Recognition
Model | Modal | Normal | Abnormal | |
|---|---|---|---|---|
RGB | Skeleton | |||
CorrNet | ✓ | × | 76.00% | 45.34% |
CSN | ✓ | × | 61.25% | 39.76% |
ResNet3D | ✓ | × | 84.20% | 58.60% |
SlowFast | ✓ | × | 90.25% | 66.12% |
ViViT | ✓ | × | 65.24% | 31.22% |
MViT | ✓ | × | 81.58% | 53.21% |
AcT | × | ✓ | 60.00% | 30.14% |
SR2M | ✓ | ✓ | 91.20% | 70.23% |
