{"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/multi-task-fusion-for-efficient-panoptic-part","title":"Multi-task Fusion for Efficient Panoptic-Part Segmentation","arxiv_id":"2212.07671","date":"2022-12-15","proceeding":null,"authors":["Sravan Kumar Jagadeesh","René Schuster","Didier Stricker"],"abstract":"In this paper, we introduce a novel network that generates semantic, instance, and part segmentation using a shared encoder and effectively fuses them to achieve panoptic-part segmentation. Unifying these three segmentation problems allows for mutually improved and consistent representation learning. To fuse the predictions of all three heads efficiently, we introduce a parameter-free joint fusion module that dynamically balances the logits and fuses them to create panoptic-part segmentation. Our method is evaluated on the Cityscapes Panoptic Parts (CPP) and Pascal Panoptic Parts (PPP) datasets. For CPP, the PartPQ of our proposed model with joint fusion surpasses the previous state-of-the-art by 1.6 and 4.7 percentage points for all areas and segments with parts, respectively. On PPP, our joint fusion outperforms a model using the previous top-down merging strategy by 3.3 percentage points in PartPQ and 10.5 percentage points in PartPQ for partitionable classes.","url_abs":"https://arxiv.org/abs/2212.07671v2","url_pdf":"https://arxiv.org/pdf/2212.07671v2.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":[],"tasks":[{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"part-level-panoptic-segmentation","task_name":"Part-aware Panoptic Segmentation"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-segmentation-on-pascal-panoptic-parts","task":"Image Segmentation","dataset":"Pascal Panoptic Parts","model":"JPPF","rank_in_archive_order":4,"of":4,"metrics":{"mIoUPartS":"54.4"},"uses_additional_data":false},{"leaderboard":"/sota/part-aware-panoptic-segmentation-on","task":"Part-aware Panoptic Segmentation","dataset":"Cityscapes Panoptic Parts","model":"JPPF","rank_in_archive_order":4,"of":4,"metrics":{"PartPQ":"61.8"},"uses_additional_data":false},{"leaderboard":"/sota/part-aware-panoptic-segmentation-on-pascal","task":"Part-aware Panoptic Segmentation","dataset":"Pascal Panoptic Parts","model":"JPPF","rank_in_archive_order":4,"of":4,"metrics":{"PartPQ":"32.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2212.07671","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}