Beyond Spatio-Temporal Priors: A Generalizable Approach for Dense Correspondence Matching
To establish identity-preserving correspondence across such transformations, we introduce FreeMatching, a generalizable framework combining generative and semantic foundation representations with heterogeneous supervision from classical datasets, tracked videos, and synthetic scenes.
Key points
- Dense correspondence matching has historically been bounded by simplifying spatio-temporal priors, such as smooth motion and rigid geometry.
- While effective for classical tasks, these assumptions break down in image editing and reference-guided generation (IEG), where transformations can preserve visual identity while breaking physical continuity.
- Teacher-guided iterative refinement further improves correspondence in IEG without dense correspondence annotations.
- Experimentally, a single FreeMatching model substantially improves correspondence quality on challenging IEG image pairs while retaining competitive performance on classical benchmarks.
Sources (2)
- [1]Beyond Spatio-Temporal Priors: A Generalizable Approach for Dense Correspondence MatchingHugging Face Daily Papers · Oct 8, 12:00 AM
To establish identity-preserving correspondence across such transformations, we introduce FreeMatching, a generalizable framework combining generative and semantic foundation representations with heterogeneous supervision from classical datasets, tracked videos, and synthetic scenes.
Dense correspondence matching has historically been bounded by simplifying spatio-temporal priors, such as smooth motion and rigid geometry.
- [2]Beyond Spatio-Temporal Priors: A Generalizable Approach for Dense Correspondence MatchingarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 05:53 PM · same content
Extractive summary: sentences quoted from the sources.