Stable Single-Pixel Contrastive Learning for Semantic and Geometric Tasks

Leonid Pogorelyuk, Niels Bracher, Aaron Verkleeren, Lars Kühmichel, Stefan T. Radev
12/4/2025
cs.CV

No confident code implementation found for this paper.

Abstract

We pilot a family of stable contrastive losses for learning pixel-level representations that jointly capture semantic and geometric information. Our approach maps each pixel of an image to an overcomplete descriptor that is both view-invariant and semantically meaningful. It enables precise point-correspondence across images without requiring momentum-based teacher-student training. Two experiments in synthetic 2D and 3D environments demonstrate the properties of our loss and the resulting overcomplete representations.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Cite this paper

@article{pogorelyuk2025stable,
  title  = {Stable Single-Pixel Contrastive Learning for Semantic and Geometric Tasks},
  author = {Leonid Pogorelyuk and Niels Bracher and Aaron Verkleeren and Lars Kühmichel and Stefan T. Radev},
  year   = {2025},
  eprint = {2512.04970},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2512.04970v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2512.04970.svg)](https://codeofpaper.com/paper/2512.04970)

Discussion