{"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/fully-convolutional-multi-class-multiple","title":"Fully Convolutional Multi-Class Multiple Instance Learning","arxiv_id":"1412.7144","date":"2014-12-22","proceeding":null,"authors":["Deepak Pathak","Evan Shelhamer","Jonathan Long","Trevor Darrell"],"abstract":"Multiple instance learning (MIL) can reduce the need for costly annotation in\ntasks such as semantic segmentation by weakening the required degree of\nsupervision. We propose a novel MIL formulation of multi-class semantic\nsegmentation learning by a fully convolutional network. In this setting, we\nseek to learn a semantic segmentation model from just weak image-level labels.\nThe model is trained end-to-end to jointly optimize the representation while\ndisambiguating the pixel-image label assignment. Fully convolutional training\naccepts inputs of any size, does not need object proposal pre-processing, and\noffers a pixelwise loss map for selecting latent instances. Our multi-class MIL\nloss exploits the further supervision given by images with multiple labels. We\nevaluate this approach through preliminary experiments on the PASCAL VOC\nsegmentation challenge.","url_abs":"http://arxiv.org/abs/1412.7144v4","url_pdf":"http://arxiv.org/pdf/1412.7144v4.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":"fully-convolutional-multi-class-multiple","repo_url":"https://github.com/ahounkanrin/FCN-MIL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"multiple-instance-learning","task_name":"Multiple Instance Learning"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"weakly-supervised-semantic-segmentation","task_name":"Weakly-Supervised Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1412.7144","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}