Research map / Video / Action recognition
Adversarial Cross-Domain Action Recognition with Co-Attention
Boxiao Pan, Zhangjie Cao, Ehsan Adeli, Juan Carlos Niebles
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use TCoN instead of CDAN | TSN · HMDB51 → UCF101 (1) | accuracy (%): 91.86 -> 96.78 | Improved |
| use TCoN instead of DAAA | C3D · UCF50 → Olympic_Sports | accuracy (%): 91.37 -> 95.92 | Improved |
| use TCoN instead of TA3N | TRN · Olympic_Sports → UCF50 | accuracy (%): 92.92 -> 96.79 | Improved |
| use TCoN instead of TA3N | TRN · UCF50 → Olympic_Sports | accuracy (%): 98.15 -> 96.82 | Hurt |
| use TCoN instead of TA3N | TRN · Jester (S) → Jester (T) | accuracy (%): 60.11 -> 62.53 | Improved |
| add a segment-level discriminator (ablation) | TCoN · Jester (S) → Jester (T) | accuracy (%): 71.13 -> 72.24 | Improved |
| use target-aligned source features for matching (ablation) | TCoN · Jester (S) → Jester (T) | accuracy (%): 65.48 -> 72.24 | Improved |
| use co-attention instead of self-attention (ablation) | TCoN · Jester (S) → Jester (T) | accuracy (%): 57.95 -> 72.24 | Improved |
| weight segment predictions with attention scores instead of averaging (ablation) | TCoN · Jester (S) → Jester (T) | accuracy (%): 63.13 -> 72.24 | Improved |
Closest papers by meaning
- CAST: Cross-Attention in Space and Time for Video Action Recognition (NeurIPS 2023 · similarity 0.86)
- Interact Before Align: Leveraging Cross-Modal Knowledge for Domain Adaptive Action Recognition (CVPR 2022 · similarity 0.86)
- Self-Paced Collaborative and Adversarial Network for Unsupervised Domain Adaptation (TPAMI 2019 · similarity 0.86)
- DDLSTM: Dual-Domain LSTM for Cross-Dataset Action Recognition (CVPR 2019 · similarity 0.85)
- CDFSL-V: Cross-Domain Few-Shot Learning for Videos (ICCV 2023 · similarity 0.85)
- Learning Causal Domain-Invariant Temporal Dynamics for Few-Shot Action Recognition (ICML 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
