{"ID":2833087,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.04970","arxiv_id":"2512.04970","title":"Stable Single-Pixel Contrastive Learning for Semantic and Geometric Tasks","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.","short_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...","url_abs":"https://arxiv.org/abs/2512.04970","url_pdf":"https://arxiv.org/pdf/2512.04970v1","authors":"[\"Leonid Pogorelyuk\",\"Niels Bracher\",\"Aaron Verkleeren\",\"Lars Kühmichel\",\"Stefan T. Radev\"]","published":"2025-12-04T16:38:26Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}