Research map / Label-efficient and robust learning / Semi-supervised learning
Semi-Supervised Learning of Semantic Correspondence With Pseudo-Labels
Jiwon Kim, Kwangrok Ryoo, Junyoung Seo, Gyuseong Lee, Daehwan Kim, Hansang Cho, Seungryong Kim
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add pseudo-label-based semi-supervised training to CATs | CATs · PF-PASCAL | PCK@0.1: 92.6 -> 93.5 | Improved |
| add pseudo-label-based semi-supervised training to CATs | CATs · PF-Willow | PCK@0.1: 79.2 -> 82.1 | Improved |
| add pseudo-label-based semi-supervised training to CATs | CATs · SPair-71k | PCK@0.1: 49.9 -> 50.7 | Improved |
| add foreground object masking to pseudo-label selection (ablation) | SemiMatch · PF-PASCAL | PCK@0.05: 73.9 -> 80.1 | Improved |
| add forward-backward consistency masking (ablation) | SemiMatch · PF-Willow | PCK@0.05: 52.9 -> 54.0 | Improved |
| add uncertainty-based threshold weighting (ablation) | SemiMatch · PF-PASCAL | PCK@0.05: 72.3 -> 80.1 | Improved |
| replace ordinary CutOut with weak keypoint-guided CutOut (ablation) | SemiMatch Base · PF-Willow | PCK@0.05: 51.8 -> 54.0 | Improved |
| use strong rather than weak keypoint-guided CutOut (ablation) | SemiMatch Base · PF-PASCAL | PCK@0.1: 93.6 -> 93.4 | No clear effect |
| remove the 20-epoch supervised warm-up (ablation) | SemiMatch · PF-PASCAL | PCK@0.1: 93.6 -> 93.5 | No clear effect |
| replace contrastive loss with AEPE loss (ablation) | SemiMatch · PF-PASCAL | PCK@0.1: 93.6 -> 92.7 | Hurt |
Closest papers by meaning
- Learning Semantic Correspondence with Sparse Annotations (ECCV 2022 · similarity 0.89)
- Weakly Supervised Learning of Semantic Correspondence through Cascaded Online Correspondence Refinement (ICCV 2023 · similarity 0.89)
- Do It Yourself: Learning Semantic Correspondence from Pseudo-Labels (ICCV 2025 · similarity 0.89)
- FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence (NeurIPS 2020 · similarity 0.88)
- Barely-Supervised Learning: Semi-supervised Learning with Very Few Labeled Images (AAAI 2022 · similarity 0.88)
- OTMatch: Improving Semi-Supervised Learning with Optimal Transport (ICML 2024 · similarity 0.88)
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.
