{"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/semantic-self-adaptation-enhancing","title":"Semantic Self-adaptation: Enhancing Generalization with a Single Sample","arxiv_id":"2208.05788","date":"2022-08-10","proceeding":null,"authors":["Sherwin Bahmani","Oliver Hahn","Eduard Zamfir","Nikita Araslanov","Daniel Cremers","Stefan Roth"],"abstract":"The lack of out-of-domain generalization is a critical weakness of deep networks for semantic segmentation. Previous studies relied on the assumption of a static model, i. e., once the training process is complete, model parameters remain fixed at test time. In this work, we challenge this premise with a self-adaptive approach for semantic segmentation that adjusts the inference process to each input sample. Self-adaptation operates on two levels. First, it fine-tunes the parameters of convolutional layers to the input image using consistency regularization. Second, in Batch Normalization layers, self-adaptation interpolates between the training and the reference distribution derived from a single test sample. Despite both techniques being well known in the literature, their combination sets new state-of-the-art accuracy on synthetic-to-real generalization benchmarks. Our empirical study suggests that self-adaptation may complement the established practice of model regularization at training time for improving deep network generalization to out-of-domain data. Our code and pre-trained models are available at https://github.com/visinf/self-adaptive.","url_abs":"https://arxiv.org/abs/2208.05788v3","url_pdf":"https://arxiv.org/pdf/2208.05788v3.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":"semantic-self-adaptation-enhancing","repo_url":"https://github.com/visinf/self-adaptive","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"domain-generalization","task_name":"Domain Generalization"},{"task_slug":"one-shot-unsupervised-domain-adaptation","task_name":"One-shot Unsupervised Domain Adaptation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"test","method_name":"Test"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/domain-generalization-on-gta-to-avg","task":"Domain Generalization","dataset":"GTA-to-Avg(Cityscapes,BDD,Mapillary)","model":"Self-adaptation (ResNet - 50)","rank_in_archive_order":23,"of":24,"metrics":{"mIoU":"44,07"},"uses_additional_data":false},{"leaderboard":"/sota/domain-generalization-on-gta-to-avg","task":"Domain Generalization","dataset":"GTA-to-Avg(Cityscapes,BDD,Mapillary)","model":"Self-adaptation (ResNet - 101)","rank_in_archive_order":24,"of":24,"metrics":{"mIoU":"44,89"},"uses_additional_data":false},{"leaderboard":"/sota/domain-generalization-on-gta5-to-cityscapes","task":"Domain Generalization","dataset":"GTA5-to-Cityscapes","model":"Self-adaptation (ResNet - 101)","rank_in_archive_order":7,"of":8,"metrics":{"mIoU":"46.99"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2208.05788","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}