{"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-image-segmentation-with-deep","title":"Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs","arxiv_id":"1412.7062","date":"2014-12-22","proceeding":null,"authors":["Liang-Chieh Chen","George Papandreou","Iasonas Kokkinos","Kevin Murphy","Alan L. Yuille"],"abstract":"Deep Convolutional Neural Networks (DCNNs) have recently shown state of the\nart performance in high level vision tasks, such as image classification and\nobject detection. This work brings together methods from DCNNs and\nprobabilistic graphical models for addressing the task of pixel-level\nclassification (also called \"semantic image segmentation\"). We show that\nresponses at the final layer of DCNNs are not sufficiently localized for\naccurate object segmentation. This is due to the very invariance properties\nthat make DCNNs good for high level tasks. We overcome this poor localization\nproperty of deep networks by combining the responses at the final DCNN layer\nwith a fully connected Conditional Random Field (CRF). Qualitatively, our\n\"DeepLab\" system is able to localize segment boundaries at a level of accuracy\nwhich is beyond previous methods. Quantitatively, our method sets the new\nstate-of-art at the PASCAL VOC-2012 semantic image segmentation task, reaching\n71.6% IOU accuracy in the test set. We show how these results can be obtained\nefficiently: Careful network re-purposing and a novel application of the 'hole'\nalgorithm from the wavelet community allow dense computation of neural net\nresponses at 8 frames per second on a modern GPU.","url_abs":"http://arxiv.org/abs/1412.7062v4","url_pdf":"http://arxiv.org/pdf/1412.7062v4.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-image-segmentation-with-deep","repo_url":"https://bitbucket.org/deeplab/deeplab-public","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/BardOfCodes/pytorch_deeplab_large_fov","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/Daeijavad/Deeplab-CRF","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/DeepMotionAIResearch/DenseASPP","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/Jasonlee1995/DeepLab_v1","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/NASA-NeMO-Net/NeMO-Net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/TheLegendAli/DeepLab-Context","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/arahusky/Tensorflow-Segmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/johnnylu305/Simple-does-it-weakly-supervised-instance-and-semantic-segmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/nightrome/cocostuff10k","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/open-cv/deeplab-v1","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/pathak22/ccnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/tensorflow/models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/tensorflow/models/tree/master/research/deeplab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/wangleihitcs/DeepLab-V1-PyTorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/2024-MindSpore-1/Code4/tree/main/DeepLabV3P","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/code-implementation1/Code9/tree/main/DeepLabV3P","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"semantic-image-segmentation-with-deep","repo_url":"https://github.com/open-mmlab/mmsegmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":null,"task_name":"GPU"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"crf","method_name":"CRF"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"deeplab","method_name":"DeepLab"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"feedforward-network","method_name":"Feedforward Network"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"sgd-with-momentum","method_name":"SGD with Momentum"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/real-time-semantic-segmentation-on-camvid","task":"Real-Time Semantic Segmentation","dataset":"CamVid","model":"DeepLab","rank_in_archive_order":27,"of":29,"metrics":{"Frame (fps)":"4.9","Time (ms)":"203","mIoU":"61.6%"},"uses_additional_data":false},{"leaderboard":"/sota/real-time-semantic-segmentation-on-cityscapes","task":"Real-Time Semantic Segmentation","dataset":"Cityscapes test","model":"DeepLab","rank_in_archive_order":37,"of":39,"metrics":{"Frame (fps)":"0.25","Time (ms)":"4000","mIoU":"63.1%"},"uses_additional_data":false},{"leaderboard":"/sota/scene-segmentation-on-sun-rgbd","task":"Scene Segmentation","dataset":"SUN-RGBD","model":"DeepLab-LargeFOV","rank_in_archive_order":3,"of":5,"metrics":{"Mean IoU":"32.08"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-camvid","task":"Semantic Segmentation","dataset":"CamVid","model":"DeepLab-MSc-CRF-LargeFOV","rank_in_archive_order":19,"of":21,"metrics":{"Mean IoU":"61.6%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-cityscapes","task":"Semantic Segmentation","dataset":"Cityscapes test","model":"DeepLab","rank_in_archive_order":98,"of":105,"metrics":{"Mean IoU (class)":"63.1%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-pascal-voc-2012","task":"Semantic Segmentation","dataset":"PASCAL VOC 2012 test","model":"DeepLab-MSc-CRF-LargeFOV (VGG-16)","rank_in_archive_order":38,"of":51,"metrics":{"Mean IoU":"71.6%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1412.7062","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1412.7062"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tensorflow/models","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nightrome/cocostuff10k","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/code-implementation1/Code9/tree/main/DeepLabV3P","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Jasonlee1995/DeepLab_v1","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Daeijavad/Deeplab-CRF","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/wangleihitcs/DeepLab-V1-PyTorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/arahusky/Tensorflow-Segmentation","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2024-MindSpore-1/Code4/tree/main/DeepLabV3P","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://bitbucket.org/deeplab/deeplab-public","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/DeepMotionAIResearch/DenseASPP","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/johnnylu305/Simple-does-it-weakly-supervised-instance-and-semantic-segmentation","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/open-mmlab/mmsegmentation","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/TheLegendAli/DeepLab-Context","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/pathak22/ccnn","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/NASA-NeMO-Net/NeMO-Net","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/BardOfCodes/pytorch_deeplab_large_fov","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tensorflow/models/tree/master/research/deeplab","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/open-cv/deeplab-v1","reach":{"status":"ok","spdx":"NOASSERTION"}}],"summary":{"unverified":1},"by_repo_kind":{"listed":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"9bf5955c8ae6f932","entry":"draw_results","repo":"arahusky/Tensorflow-Segmentation","repo_kind":"listed","path":"convolutional_autoencoder.py","file_url":"https://github.com/arahusky/Tensorflow-Segmentation/blob/HEAD/convolutional_autoencoder.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9bf5955c8ae6f932"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}