{"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/reseg-a-recurrent-neural-network-based-model","title":"ReSeg: A Recurrent Neural Network-based Model for Semantic Segmentation","arxiv_id":"1511.07053","date":"2015-11-22","proceeding":null,"authors":["Francesco Visin","Marco Ciccone","Adriana Romero","Kyle Kastner","Kyunghyun Cho","Yoshua Bengio","Matteo Matteucci","Aaron Courville"],"abstract":"We propose a structured prediction architecture, which exploits the local\ngeneric features extracted by Convolutional Neural Networks and the capacity of\nRecurrent Neural Networks (RNN) to retrieve distant dependencies. The proposed\narchitecture, called ReSeg, is based on the recently introduced ReNet model for\nimage classification. We modify and extend it to perform the more challenging\ntask of semantic segmentation. Each ReNet layer is composed of four RNN that\nsweep the image horizontally and vertically in both directions, encoding\npatches or activations, and providing relevant global information. Moreover,\nReNet layers are stacked on top of pre-trained convolutional layers, benefiting\nfrom generic local features. Upsampling layers follow ReNet layers to recover\nthe original image resolution in the final predictions. The proposed ReSeg\narchitecture is efficient, flexible and suitable for a variety of semantic\nsegmentation tasks. We evaluate ReSeg on several widely-used semantic\nsegmentation datasets: Weizmann Horse, Oxford Flower, and CamVid; achieving\nstate-of-the-art performance. Results show that ReSeg can act as a suitable\narchitecture for semantic segmentation tasks, and may have further applications\nin other structured prediction problems. The source code and model\nhyperparameters are available on https://github.com/fvisin/reseg.","url_abs":"http://arxiv.org/abs/1511.07053v3","url_pdf":"http://arxiv.org/pdf/1511.07053v3.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":"reseg-a-recurrent-neural-network-based-model","repo_url":"https://github.com/fvisin/reseg","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"reseg-a-recurrent-neural-network-based-model","repo_url":"https://github.com/SConsul/ReSeg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"reseg-a-recurrent-neural-network-based-model","repo_url":"https://github.com/MindCode-4/code-13/tree/main/ReSeg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"reseg-a-recurrent-neural-network-based-model","repo_url":"https://github.com/MindCode-4/code-8/tree/main/ReSeg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"reseg-a-recurrent-neural-network-based-model","repo_url":"https://github.com/mindspore-ai/contrib/tree/master/application/ReSeg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"structured-prediction","task_name":"Structured Prediction"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-segmentation-on-camvid","task":"Semantic Segmentation","dataset":"CamVid","model":"ReSeg","rank_in_archive_order":20,"of":21,"metrics":{"Global Accuracy":"88.7%","Mean IoU":"58.8%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1511.07053","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}