{"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/autofocus-layer-for-semantic-segmentation","title":"Autofocus Layer for Semantic Segmentation","arxiv_id":"1805.08403","date":"2018-05-22","proceeding":null,"authors":["Yao Qin","Konstantinos Kamnitsas","Siddharth Ancha","Jay Nanavati","Garrison Cottrell","Antonio Criminisi","Aditya Nori"],"abstract":"We propose the autofocus convolutional layer for semantic segmentation with\nthe objective of enhancing the capabilities of neural networks for multi-scale\nprocessing. Autofocus layers adaptively change the size of the effective\nreceptive field based on the processed context to generate more powerful\nfeatures. This is achieved by parallelising multiple convolutional layers with\ndifferent dilation rates, combined by an attention mechanism that learns to\nfocus on the optimal scales driven by context. By sharing the weights of the\nparallel convolutions we make the network scale-invariant, with only a modest\nincrease in the number of parameters. The proposed autofocus layer can be\neasily integrated into existing networks to improve a model's representational\npower. We evaluate our models on the challenging tasks of multi-organ\nsegmentation in pelvic CT and brain tumor segmentation in MRI and achieve very\npromising performance.","url_abs":"http://arxiv.org/abs/1805.08403v3","url_pdf":"http://arxiv.org/pdf/1805.08403v3.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":"autofocus-layer-for-semantic-segmentation","repo_url":"https://github.com/yaq007/Autofocus-Layer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"autofocus-layer-for-semantic-segmentation","repo_url":"https://github.com/luvgold/auotofoucus3D-Brats","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"autofocus-layer-for-semantic-segmentation","repo_url":"https://github.com/perslev/Autofocus-Layer-TF","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"brain-tumor-segmentation","task_name":"Brain Tumor Segmentation"},{"task_slug":"medical-image-segmentation","task_name":"Medical Image Segmentation"},{"task_slug":"organ-segmentation","task_name":"Organ Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"tumor-segmentation","task_name":"Tumor Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/brain-tumor-segmentation-on-brats-2015","task":"Brain Tumor Segmentation","dataset":"BRATS-2015","model":"AFN-6","rank_in_archive_order":4,"of":4,"metrics":{"Dice Score":"84%"},"uses_additional_data":true}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}