{"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/diffmatch-visual-language-guidance-makes","title":"SemiCD-VL: Visual-Language Model Guidance Makes Better Semi-supervised Change Detector","arxiv_id":"2405.04788","date":"2024-05-08","proceeding":null,"authors":["Kaiyu Li","Xiangyong Cao","Yupeng Deng","Jiayi Song","Junmin Liu","Deyu Meng","Zhi Wang"],"abstract":"Change Detection (CD) aims to identify pixels with semantic changes between images. However, annotating massive numbers of pixel-level images is labor-intensive and costly, especially for multi-temporal images, which require pixel-wise comparisons by human experts. Considering the excellent performance of visual language models (VLMs) for zero-shot, open-vocabulary, etc. with prompt-based reasoning, it is promising to utilize VLMs to make better CD under limited labeled data. In this paper, we propose a VLM guidance-based semi-supervised CD method, namely SemiCD-VL. The insight of SemiCD-VL is to synthesize free change labels using VLMs to provide additional supervision signals for unlabeled data. However, almost all current VLMs are designed for single-temporal images and cannot be directly applied to bi- or multi-temporal images. Motivated by this, we first propose a VLM-based mixed change event generation (CEG) strategy to yield pseudo labels for unlabeled CD data. Since the additional supervised signals provided by these VLM-driven pseudo labels may conflict with the pseudo labels from the consistency regularization paradigm (e.g. FixMatch), we propose the dual projection head for de-entangling different signal sources. Further, we explicitly decouple the bi-temporal images semantic representation through two auxiliary segmentation decoders, which are also guided by VLM. Finally, to make the model more adequately capture change representations, we introduce metric-aware supervision by feature-level contrastive loss in auxiliary branches. Extensive experiments show the advantage of SemiCD-VL. For instance, SemiCD-VL improves the FixMatch baseline by +5.3 IoU on WHU-CD and by +2.4 IoU on LEVIR-CD with 5% labels. In addition, our CEG strategy, in an un-supervised manner, can achieve performance far superior to state-of-the-art un-supervised CD methods.","url_abs":"https://arxiv.org/abs/2405.04788v5","url_pdf":"https://arxiv.org/pdf/2405.04788v5.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":"diffmatch-visual-language-guidance-makes","repo_url":"https://github.com/likyoo/diffmatch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"diffmatch-visual-language-guidance-makes","repo_url":"https://github.com/likyoo/semicd-vl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"change-detection","task_name":"Change Detection"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"semi-supervised-change-detection","task_name":"Semi-supervised Change Detection"}],"methods":[{"method_slug":"fixmatch","method_name":"FixMatch"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semi-supervised-change-detection-on-levir-cd-1","task":"Semi-supervised Change Detection","dataset":"LEVIR-CD - 10% labeled data","model":"DiffMatch","rank_in_archive_order":3,"of":5,"metrics":{"IoU":"82.6"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-levir-cd","task":"Semi-supervised Change Detection","dataset":"LEVIR-CD - 5% labeled data","model":"DiffMatch","rank_in_archive_order":2,"of":5,"metrics":{"IoU":"81.9"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-whu-10","task":"Semi-supervised Change Detection","dataset":"WHU - 10% labeled data","model":"DiffMatch","rank_in_archive_order":1,"of":4,"metrics":{"IoU":"83.2"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-whu-5","task":"Semi-supervised Change Detection","dataset":"WHU - 5% labeled data","model":"DiffMatch","rank_in_archive_order":1,"of":4,"metrics":{"IoU":"81.8"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}