{"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/pixelwise-instance-segmentation-with-a","title":"Pixelwise Instance Segmentation with a Dynamically Instantiated Network","arxiv_id":"1704.02386","date":"2017-04-07","proceeding":"CVPR 2017 7","authors":["Anurag Arnab","Philip H. S. Torr"],"abstract":"Semantic segmentation and object detection research have recently achieved\nrapid progress. However, the former task has no notion of different instances\nof the same object, and the latter operates at a coarse, bounding-box level. We\npropose an Instance Segmentation system that produces a segmentation map where\neach pixel is assigned an object class and instance identity label. Most\napproaches adapt object detectors to produce segments instead of boxes. In\ncontrast, our method is based on an initial semantic segmentation module, which\nfeeds into an instance subnetwork. This subnetwork uses the initial\ncategory-level segmentation, along with cues from the output of an object\ndetector, within an end-to-end CRF to predict instances. This part of our model\nis dynamically instantiated to produce a variable number of instances per\nimage. Our end-to-end approach requires no post-processing and considers the\nimage holistically, instead of processing independent proposals. Therefore,\nunlike some related work, a pixel cannot belong to multiple instances.\nFurthermore, far more precise segmentations are achieved, as shown by our\nstate-of-the-art results (particularly at high IoU thresholds) on the Pascal\nVOC and Cityscapes datasets.","url_abs":"http://arxiv.org/abs/1704.02386v1","url_pdf":"http://arxiv.org/pdf/1704.02386v1.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":"pixelwise-instance-segmentation-with-a","repo_url":"https://github.com/hmph/dynamically-instantiated-network","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"panoptic-segmentation","task_name":"Panoptic Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"crf","method_name":"CRF"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/instance-segmentation-on-cityscapes","task":"Instance Segmentation","dataset":"Cityscapes test","model":"Dynamically Instantiated Network","rank_in_archive_order":2,"of":11,"metrics":{},"uses_additional_data":false},{"leaderboard":"/sota/panoptic-segmentation-on-cityscapes-test","task":"Panoptic Segmentation","dataset":"Cityscapes test","model":"Dynamically Instantiated Network","rank_in_archive_order":10,"of":10,"metrics":{"PQ":"55.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.02386","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}