Zero Shot Skeletal Action Recognition On Ntu 1
评估指标
Accuracy (10 unseen classes)
Accuracy (24 unseen classes)
评测结果
各个模型在此基准测试上的表现结果
模型名称 | Accuracy (10 unseen classes) | Accuracy (24 unseen classes) | Paper Title | Repository |
|---|---|---|---|---|
| PURLS | 71.95 | 52.01 | Part-aware Unified Representation of Language and Skeleton for Zero-shot Action Recognition | - |
| STAR | 63.30 | 44.30 | Fine-Grained Side Information Guided Dual-Prompts for Zero-Shot Skeleton Action Recognition | - |
| TDSM | 74.15 | 65.06 | TDSM: Triplet Diffusion for Skeleton-Text Matching in Zero-Shot Action Recognition | - |
| SMIE | 65.74 | 45.30 | Zero-shot Skeleton-based Action Recognition via Mutual Information Estimation and Maximization | - |
| SA-DVAE | 68.77 | 46.12 | SA-DVAE: Improving Zero-Shot Skeleton-Based Action Recognition by Disentangled Variational Autoencoders | - |
| SynSE | 62.69 | 38.70 | Syntactically Guided Generative Embeddings for Zero-Shot Skeleton Action Recognition | - |
| CADA-VAE | 59.53 | 35.77 | Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders | - |
| SA-DVAE + augmented text | - | - | SA-DVAE: Improving Zero-Shot Skeleton-Based Action Recognition by Disentangled Variational Autoencoders | - |
0 of 8 row(s) selected.