AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Source-Free Video Domain Adaptation by Learning Temporal Consistency for Action Recognition

Yuecong Xu, Jianfei Yang, Haozhi Cao, Keyu Wu, Min Wu, Zhenghua Chen

ECCV 2022 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Source-free domain adaptation · 39 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
add source prediction consistency to feature consistency (FC only -> full) (ablation)ATCoN · UCF-HMDB-fulltop-1 accuracy: 77.78 -> 79.72Improved
add feature consistency to source prediction consistency (ablation)ATCoN · UCF-HMDB-fulltop-1 accuracy: 77.50 -> 78.89Improved
extend source prediction consistency to the overall target prediction (ablation)ATCoN · UCF-HMDB-fulltop-1 accuracy: 76.67 -> 77.50Improved
add feature consistency plus source prediction consistency to source-only (ablation)ATCoN · UCF-HMDB-fulltop-1 accuracy: 72.78 -> 78.89Improved
add IM loss and pseudo-labeling on top of temporal consistency (ablation)ATCoN-TC · UCF-HMDB-fulltop-1 accuracy: 78.89 -> 79.72Improved
add IM loss and pseudo-labeling on top of temporal consistency (H51->U101) (ablation)ATCoN-TC · UCF-HMDB-fulltop-1 accuracy: 84.59 -> 85.29Improved
add local weight module (any form) (ablation)ATCoN-NA · UCF-HMDB-fulltop-1 accuracy: 78.33 -> 79.72Improved
apply local relevance weight to both feature and prediction (vs feature only) (ablation)ATCoN-A@F · UCF-HMDB-fulltop-1 accuracy: 79.17 -> 79.72Improved
apply local relevance weight to both feature and prediction (vs prediction only) (ablation)ATCoN-A@P · UCF-HMDB-fulltop-1 accuracy: 84.41 -> 85.29Improved
add local weight module (H51->U101) (ablation)ATCoN-NA · UCF-HMDB-fulltop-1 accuracy: 83.89 -> 85.29Improved
use ATCoN instead of image-based SFDA methods (SHOT, MA, CPGA)source-free video adaptation · Daily-DAtop-1 accuracy: 26.46 -> 33.53Improved
apply image-based SFDA method SHOT instead of source-onlyTRN · Daily-DAtop-1 accuracy: 27.89 -> 25.30Hurt

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