AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Interact Before Align: Leveraging Cross-Modal Knowledge for Domain Adaptive Action Recognition

Lijin Yang, Yifei Huang, Yusuke Sugano, Yoichi Sato

CVPR 2022 · Vision-language and multimodal › Cross-modal learning · approach: Unimodal-to-multimodal representation transfer · 36 citations (OpenAlex)

Check 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 changedWhereMeasuredEffect
Use CIA instead of Kim et al.I3D multi-modal domain adaptation · UCF-HMDB U→Hclass-wise action recognition accuracy: 84.70 -> 88.33Improved
Use CIA instead of Kim et al.I3D multi-modal domain adaptation · UCF-HMDB H→Uclass-wise action recognition accuracy: 92.80 -> 94.05Improved
Add CIA to TA3NI3D-TRN · UCF-HMDB U→Hclass-wise action recognition accuracy: 89.17 -> 91.94Improved
Use CIA instead of STCDAEPIC-Kitchens-55 · mean across six domain transfersclass-wise action recognition accuracy: 51.2 -> 52.2Improved
Use CIA with RGB and Flow instead of MM-SADAEPIC-Kitchens-100 · validationaction accuracy: 12.80 -> 14.25Improved
Add both self-refinement and cross-refinement (ablation)CIA on EPIC-Kitchens-100 · validationaction accuracy: 19.19 -> 20.49Improved
Add self-refinement (ablation)CIA on EPIC-Kitchens-100 · validationaction accuracy: 19.19 -> 19.56Improved
Add cross-refinement (ablation)CIA on EPIC-Kitchens-100 · validationaction accuracy: 19.19 -> 19.67Improved
Add SC to MC (ablation)CIA on EPIC-Kitchens-100 · validationaction accuracy: 19.98 -> 20.49Improved
Add MC to SC (ablation)CIA on EPIC-Kitchens-100 · validationaction accuracy: 19.83 -> 20.49Improved
Use multi-scale rather than single-scale correlation (ablation)SC on EPIC-Kitchens-100 · validation, domain adaptationaction accuracy: 19.62 -> 19.83Improved
Replace SC with spatial attention (ablation)EPIC-Kitchens-100 domain adaptation · validationaction accuracy: 19.83 -> 19.39Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy