{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/correlation-loss-enforcing-correlation","title":"Correlation Loss: Enforcing Correlation between Classification and Localization","arxiv_id":"2301.01019","date":"2023-01-03","proceeding":null,"authors":["Fehmi Kahraman","Kemal Oksuz","Sinan Kalkan","Emre Akbas"],"abstract":"Object detectors are conventionally trained by a weighted sum of classification and localization losses. Recent studies (e.g., predicting IoU with an auxiliary head, Generalized Focal Loss, Rank & Sort Loss) have shown that forcing these two loss terms to interact with each other in non-conventional ways creates a useful inductive bias and improves performance. Inspired by these works, we focus on the correlation between classification and localization and make two main contributions: (i) We provide an analysis about the effects of correlation between classification and localization tasks in object detectors. We identify why correlation affects the performance of various NMS-based and NMS-free detectors, and we devise measures to evaluate the effect of correlation and use them to analyze common detectors. (ii) Motivated by our observations, e.g., that NMS-free detectors can also benefit from correlation, we propose Correlation Loss, a novel plug-in loss function that improves the performance of various object detectors by directly optimizing correlation coefficients: E.g., Correlation Loss on Sparse R-CNN, an NMS-free method, yields 1.6 AP gain on COCO and 1.8 AP gain on Cityscapes dataset. Our best model on Sparse R-CNN reaches 51.0 AP without test-time augmentation on COCO test-dev, reaching state-of-the-art. Code is available at https://github.com/fehmikahraman/CorrLoss","url_abs":"https://arxiv.org/abs/2301.01019v1","url_pdf":"https://arxiv.org/pdf/2301.01019v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"correlation-loss-enforcing-correlation","repo_url":"https://github.com/fehmikahraman/corrloss","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"inductive-bias","task_name":"Inductive Bias"},{"task_slug":"object","task_name":"Object"}],"methods":[{"method_slug":"focal-loss","method_name":"Focal Loss"},{"method_slug":"generalized-focal-loss","method_name":"Generalized Focal Loss"},{"method_slug":"sparse-r-cnn","method_name":"Sparse R-CNN"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2301.01019","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}