{"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/leveraging-out-of-distribution-unlabeled","title":"Leveraging Out-of-Distribution Unlabeled Images: Semi-Supervised Semantic Segmentation with an Open-Vocabulary Model","arxiv_id":null,"date":"2025-07-04","proceeding":null,"authors":["WooSeok Shin","Jisu Kang","Hyeonki Jeong","Jin Sob Kim","Sung Won Han"],"abstract":"In semi-supervised semantic segmentation, existing studies have shown promising results in academic settings with controlled splits of benchmark datasets. However, the potential benefits of leveraging significantly larger sets of unlabeled images remain unexplored. In real-world scenarios, abundant unlabeled images are often available from online sources (web-scraped images) or large-scale datasets. However, these images may have different distributions from those of the target dataset, a situation known as out-of-distribution (OOD). Using these images as unlabeled data in semi-supervised learning can lead to inaccurate pseudo-labels, potentially misguiding network training. In this paper, we propose a new semi-supervised semantic segmentation framework with an open-vocabulary segmentation model (SemiOVS) to effectively utilize unlabeled OOD images. Extensive experiments on Pascal VOC and Context datasets demonstrate two key findings: (1) using additional unlabeled images improves the performance of semi-supervised learners in scenarios with few labels, and (2) using the open-vocabulary segmentation (OVS) model to pseudo-label OOD images leads to substantial performance gains. In particular, SemiOVS outperforms existing PrevMatch and SemiVL methods by +3.5 and +3.0 mIoU, respectively, on Pascal VOC with a 92-label setting, achieving state-of-the-art performance. These findings demonstrate that our approach effectively utilizes abundant unlabeled OOD images for semantic segmentation tasks. We hope this work can inspire future research and real-world applications. The code is available at https://github.com/wooseok-shin/SemiOVS","url_abs":"https://arxiv.org/abs/2507.03302","url_pdf":"https://arxiv.org/pdf/2507.03302","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":"leveraging-out-of-distribution-unlabeled","repo_url":"https://github.com/wooseok-shin/SemiOVS","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"pseudo-label","task_name":"Pseudo Label"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"semi-supervised-semantic-segmentation","task_name":"Semi-Supervised Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-10","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 1464 labels","model":"SemiOVS (w/ SemiVL, ViT-B/16)","rank_in_archive_order":2,"of":17,"metrics":{"Validation mIoU":"88.0"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-10","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 1464 labels","model":"SemiOVS (w/ UniMatch, ResNet-101)","rank_in_archive_order":7,"of":17,"metrics":{"Validation mIoU":"81.8"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-28","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 183 labeled","model":"SemiOVS (w/ SemiVL, ViT-B/16)","rank_in_archive_order":2,"of":16,"metrics":{"Validation mIoU":"87.3"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-28","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 183 labeled","model":"SemiOVS (w/ UniMatch, ResNet-101)","rank_in_archive_order":4,"of":16,"metrics":{"Validation mIoU":"81.3"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-28","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 183 labeled","model":"SemiOVS (w/ UniMatch, ResNet-50)","rank_in_archive_order":6,"of":16,"metrics":{"Validation mIoU":"78.7"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-9","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 25% labeled","model":"SemiOVS (w/ PrevMatch, ResNet-101)","rank_in_archive_order":1,"of":27,"metrics":{"Validation mIoU":"81.3"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-9","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 25% labeled","model":"SemiOVS (w/ UniMatch, ResNet-101)","rank_in_archive_order":3,"of":27,"metrics":{"Validation mIoU":"81.0"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-29","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 366 labeled","model":"SemiOVS (w/ SemiVL, ViT-B/16)","rank_in_archive_order":2,"of":15,"metrics":{"Validation mIoU":"87.5"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-29","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 366 labeled","model":"SemiOVS (w/ UniMatch, ResNet-101)","rank_in_archive_order":5,"of":15,"metrics":{"Validation mIoU":"81.6"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-30","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 732 labeled","model":"SemiOVS (w/ SemiVL, ViT-B/16)","rank_in_archive_order":2,"of":16,"metrics":{"Validation mIoU":"87.9"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-30","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 732 labeled","model":"SemiOVS (w/ UniMatch, ResNet-101)","rank_in_archive_order":5,"of":16,"metrics":{"Validation mIoU":"81.7"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-27","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 92 labeled","model":"SemiOVS (w/ SemiVL, ViT-B/16)","rank_in_archive_order":1,"of":17,"metrics":{"Validation mIoU":"87.0"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-27","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 92 labeled","model":"SemiOVS (w/ UniMatch, ResNet-101)","rank_in_archive_order":4,"of":17,"metrics":{"Validation mIoU":"80.4"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-27","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 92 labeled","model":"SemiOVS (w/ UniMatch, ResNet-50)","rank_in_archive_order":5,"of":17,"metrics":{"Validation mIoU":"78.2"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-4","task":"Semi-Supervised Semantic Segmentation","dataset":"Pascal VOC 2012 12.5% labeled","model":"SemiOVS (w/ UniMatch, ResNet-101)","rank_in_archive_order":1,"of":38,"metrics":{"Validation mIoU":"82.4"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-4","task":"Semi-Supervised Semantic Segmentation","dataset":"Pascal VOC 2012 12.5% labeled","model":"SemiOVS (w/ PrevMatch, ResNet-101)","rank_in_archive_order":2,"of":38,"metrics":{"Validation mIoU":"82.3"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-21","task":"Semi-Supervised Semantic Segmentation","dataset":"Pascal VOC 2012 6.25% labeled","model":"SemiOVS (w/ PrevMatch, ResNet-101)","rank_in_archive_order":1,"of":19,"metrics":{"Validation mIoU":"82.0"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-21","task":"Semi-Supervised Semantic Segmentation","dataset":"Pascal VOC 2012 6.25% labeled","model":"SemiOVS (w/ UniMatch, ResNet-101)","rank_in_archive_order":3,"of":19,"metrics":{"Validation mIoU":"81.5"},"uses_additional_data":true}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}