{"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/shuffleseg-real-time-semantic-segmentation","title":"ShuffleSeg: Real-time Semantic Segmentation Network","arxiv_id":"1803.03816","date":"2018-03-10","proceeding":null,"authors":["Mostafa Gamal","Mennatullah Siam","Moemen Abdel-Razek"],"abstract":"Real-time semantic segmentation is of significant importance for mobile and\nrobotics related applications. We propose a computationally efficient\nsegmentation network which we term as ShuffleSeg. The proposed architecture is\nbased on grouped convolution and channel shuffling in its encoder for improving\nthe performance. An ablation study of different decoding methods is compared\nincluding Skip architecture, UNet, and Dilation Frontend. Interesting insights\non the speed and accuracy tradeoff is discussed. It is shown that skip\narchitecture in the decoding method provides the best compromise for the goal\nof real-time performance, while it provides adequate accuracy by utilizing\nhigher resolution feature maps for a more accurate segmentation. ShuffleSeg is\nevaluated on CityScapes and compared against the state of the art real-time\nsegmentation networks. It achieves 2x GFLOPs reduction, while it provides on\npar mean intersection over union of 58.3% on CityScapes test set. ShuffleSeg\nruns at 15.7 frames per second on NVIDIA Jetson TX2, which makes it of great\npotential for real-time applications.","url_abs":"http://arxiv.org/abs/1803.03816v2","url_pdf":"http://arxiv.org/pdf/1803.03816v2.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":"shuffleseg-real-time-semantic-segmentation","repo_url":"https://github.com/MSiam/TFSegmentation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"shuffleseg-real-time-semantic-segmentation","repo_url":"https://github.com/Davidnet/TFSegmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"real-time-semantic-segmentation","task_name":"Real-Time Semantic Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"grouped-convolution","method_name":"Grouped Convolution"},{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.03816","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}