AI Paper Atlas

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

Active Exploration of Multimodal Complementarity for Few-Shot Action Recognition

Yuyang Wanyan, Xiaoshan Yang, Chaofan Chen, Changsheng Xu

CVPR 2023 · Vision-language and multimodal › Cross-modal learning · approach: Multi-modality image fusion · 54 citations (OpenAlex)

Read the paper

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 AMFAR's active modality selection, mutual distillation, and adaptive inference instead of the strongest listed baselinefew-shot action recognition · Kinetics 5-way 1-shotaccuracy: 74.6 -> 80.1Improved
Use AMFAR's active modality selection, mutual distillation, and adaptive inference instead of the strongest listed baselinefew-shot action recognition · Kinetics 5-way 5-shotaccuracy: 87.4 -> 92.6Improved
Use AMFAR's active modality selection, mutual distillation, and adaptive inference instead of the strongest listed baselinefew-shot action recognition · SSv2 5-way 1-shotaccuracy: 54.3 -> 61.7Improved
Use AMFAR's active modality selection, mutual distillation, and adaptive inference instead of the strongest listed baselinefew-shot action recognition · SSv2 5-way 5-shotaccuracy: 70.4 -> 79.5Improved
Use AMFAR's active modality selection, mutual distillation, and adaptive inference instead of the strongest listed baselinefew-shot action recognition · HMDB51 5-way 1-shotaccuracy: 60.3 -> 73.9Improved
Use AMFAR's active modality selection, mutual distillation, and adaptive inference instead of the strongest listed baselinefew-shot action recognition · HMDB51 5-way 5-shotaccuracy: 81.3 -> 87.8Improved
Use AMFAR's active modality selection, mutual distillation, and adaptive inference instead of the strongest listed baselinefew-shot action recognition · UCF101 5-way 1-shotaccuracy: 87.4 -> 91.2Improved
Use AMFAR's active modality selection, mutual distillation, and adaptive inference instead of the strongest listed baselinefew-shot action recognition · UCF101 5-way 5-shotaccuracy: 98.4 -> 99.0Improved
Add Active Sample Selection to group queries by the more reliable modality (ablation)AMFAR · Kinetics 5-way 1-shotaccuracy: 72.9 -> 80.1Improved
Add Active Mutual Distillation between modalities (ablation)AMFAR · SSv2 5-way 1-shotaccuracy: 58.8 -> 61.7Improved
Add Adaptive Multimodal Inference with reliability-weighted fusion (ablation)AMFAR · SSv2 5-way 5-shotaccuracy: 78.2 -> 79.5Improved
Use absolute certainty to estimate modality reliability (ablation)ASS in AMFAR · Kinetics 5-way 1-shotaccuracy: 72.9 -> 80.1Improved

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