{"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/wider-or-deeper-revisiting-the-resnet-model","title":"Wider or Deeper: Revisiting the ResNet Model for Visual Recognition","arxiv_id":"1611.10080","date":"2016-11-30","proceeding":null,"authors":["Zifeng Wu","Chunhua Shen","Anton Van Den Hengel"],"abstract":"The trend towards increasingly deep neural networks has been driven by a\ngeneral observation that increasing depth increases the performance of a\nnetwork. Recently, however, evidence has been amassing that simply increasing\ndepth may not be the best way to increase performance, particularly given other\nlimitations. Investigations into deep residual networks have also suggested\nthat they may not in fact be operating as a single deep network, but rather as\nan ensemble of many relatively shallow networks. We examine these issues, and\nin doing so arrive at a new interpretation of the unravelled view of deep\nresidual networks which explains some of the behaviours that have been observed\nexperimentally. As a result, we are able to derive a new, shallower,\narchitecture of residual networks which significantly outperforms much deeper\nmodels such as ResNet-200 on the ImageNet classification dataset. We also show\nthat this performance is transferable to other problem domains by developing a\nsemantic segmentation approach which outperforms the state-of-the-art by a\nremarkable margin on datasets including PASCAL VOC, PASCAL Context, and\nCityscapes. The architecture that we propose thus outperforms its comparators,\nincluding very deep ResNets, and yet is more efficient in memory use and\nsometimes also in training time. The code and models are available at\nhttps://github.com/itijyou/ademxapp","url_abs":"http://arxiv.org/abs/1611.10080v1","url_pdf":"http://arxiv.org/pdf/1611.10080v1.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":"wider-or-deeper-revisiting-the-resnet-model","repo_url":"https://github.com/itijyou/ademxapp","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"mxnet","reach":null},{"paper_slug":"wider-or-deeper-revisiting-the-resnet-model","repo_url":"https://github.com/ZhaoJ9014/shallower-wider-ResNet-Model-A-SW-ResNet-A","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"wider-or-deeper-revisiting-the-resnet-model","repo_url":"https://github.com/halbielee/EPS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-segmentation-on-cityscapes","task":"Semantic Segmentation","dataset":"Cityscapes test","model":"ResNet-38","rank_in_archive_order":59,"of":105,"metrics":{"Mean IoU (class)":"78.4%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-pascal-context","task":"Semantic Segmentation","dataset":"PASCAL Context","model":"ResNet-38","rank_in_archive_order":52,"of":66,"metrics":{"mIoU":"48.1"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-pascal-voc-2012","task":"Semantic Segmentation","dataset":"PASCAL VOC 2012 test","model":"ResNet-38 MS COCO","rank_in_archive_order":11,"of":51,"metrics":{"Mean IoU":"84.9%"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1611.10080","atlas_url":"https://app.syntology.ai/?focus=1611.10080","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1611.10080"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/ZhaoJ9014/shallower-wider-ResNet-Model-A-SW-ResNet-A","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/halbielee/EPS","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/itijyou/ademxapp","reach":null}],"summary":{"unverified":2},"by_repo_kind":{"official":{"samples":2,"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":2,"samples":[{"code_sha256_prefix":"462aaac067f7f126","entry":"parse_model_label","repo":"itijyou/ademxapp","repo_kind":"official","path":"iclass/ilsvrc.py","file_url":"https://github.com/itijyou/ademxapp/blob/HEAD/iclass/ilsvrc.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"462aaac067f7f126"}},{"code_sha256_prefix":"38528353018213f8","entry":"parse_model_label","repo":"itijyou/ademxapp","repo_kind":"official","path":"issegm/voc.py","file_url":"https://github.com/itijyou/ademxapp/blob/HEAD/issegm/voc.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"38528353018213f8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}