AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Cross-Model Pseudo-Labeling for Semi-Supervised Action Recognition

Yinghao Xu, Fangyun Wei, Xiao Sun, Ceyuan Yang, Yujun Shen, Bo Dai, Bolei Zhou, Stephen Lin

CVPR 2022 · Label-efficient and robust learning › Semi-supervised learning · approach: Confidence-based pseudo-label selection · 63 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 cross-model pseudo-labeling instead of FixMatch self-labeling3D-ResNet50 CMPL · Kinetics-400, 1% labeledTop-1 accuracy: 8.6 -> 17.6Improved
Use cross-model pseudo-labeling instead of FixMatch self-labeling3D-ResNet50 CMPL · UCF-101, 1% labeledTop-1 accuracy: 14.8 -> 25.1Improved
Add auxiliary cross-model supervision to FixMatch (ablation)3D-ResNet50 on Kinetics-400 · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 6.78 -> 12.04Improved
Add different frame rates and temporal locations (ablation)CMPL with auxiliary supervision · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 12.04 -> 13.71Improved
Use a different-architecture auxiliary network rather than the identical primary architecture (ablation)CMPL · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 6.78 -> 9.97Improved
Use 3D-ResNet50 × 1/4 as the auxiliary network instead of 3D-ResNet50 × 1/2 (ablation)CMPL · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 11.57 -> 12.90Improved
Increase auxiliary sampling from 8 × 8 to 16 × 4 (ablation)CMPL · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 12.04 -> 12.90Improved
Increase auxiliary sampling from 16 × 4 to 32 × 2 (ablation)CMPL · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 12.90 -> 12.94Improved
Increase temporal offset from t_s 1 to t_s 2 (ablation)CMPL · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 12.90 -> 12.64Hurt
Increase temporal offset from t_s 2 to t_s 10 (ablation)CMPL · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 12.64 -> 13.71Improved
Add both supervised and unsupervised losses to the auxiliary network (ablation)CMPL · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 6.78 -> 12.90Improved
Average the two networks' predictions rather than use cross-model pseudo-labeling (ablation)CMPL pseudo-labeling · Kinetics-400, 1% labeled, 50 epochsTop-1 accuracy: 12.90 -> 12.16Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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