{"ID":2898691,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.02493","arxiv_id":"2507.02493","title":"Temporally-Aware Supervised Contrastive Learning for Polyp Counting in Colonoscopy","abstract":"Automated polyp counting in colonoscopy is a crucial step toward automated procedure reporting and quality control, aiming to enhance the cost-effectiveness of colonoscopy screening. Counting polyps in a procedure involves detecting and tracking polyps, and then clustering tracklets that belong to the same polyp entity. Existing methods for polyp counting rely on self-supervised learning and primarily leverage visual appearance, neglecting temporal relationships in both tracklet feature learning and clustering stages. In this work, we introduce a paradigm shift by proposing a supervised contrastive loss that incorporates temporally-aware soft targets. Our approach captures intra-polyp variability while preserving inter-polyp discriminability, leading to more robust clustering. Additionally, we improve tracklet clustering by integrating a temporal adjacency constraint, reducing false positive re-associations between visually similar but temporally distant tracklets. We train and validate our method on publicly available datasets and evaluate its performance with a leave-one-out cross-validation strategy. Results demonstrate a 2.2x reduction in fragmentation rate compared to prior approaches. Our results highlight the importance of temporal awareness in polyp counting, establishing a new state-of-the-art. Code is available at https://github.com/lparolari/temporally-aware-polyp-counting.","short_abstract":"Automated polyp counting in colonoscopy is a crucial step toward automated procedure reporting and quality control, aiming to enhance the cost-effectiveness of colonoscopy screening. Counting polyps in a procedure involves detecting and tracking polyps, and then clustering tracklets that belong to the same polyp entity...","url_abs":"https://arxiv.org/abs/2507.02493","url_pdf":"https://arxiv.org/pdf/2507.02493v1","authors":"[\"Luca Parolari\",\"Andrea Cherubini\",\"Lamberto Ballan\",\"Carlo Biffi\"]","published":"2025-07-03T09:55:48Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[]","has_code":false,"code_links":[{"ID":612424,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2898691,"paper_url":"https://arxiv.org/abs/2507.02493","paper_title":"Temporally-Aware Supervised Contrastive Learning for Polyp Counting in Colonoscopy","repo_url":"https://github.com/lparolari/temporally-aware-polyp-counting","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}