{"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/global-second-order-pooling-convolutional","title":"Global Second-order Pooling Convolutional Networks","arxiv_id":"1811.12006","date":"2018-11-29","proceeding":"CVPR 2019 6","authors":["Zilin Gao","Jiangtao Xie","Qilong Wang","Peihua Li"],"abstract":"Deep Convolutional Networks (ConvNets) are fundamental to, besides\nlarge-scale visual recognition, a lot of vision tasks. As the primary goal of\nthe ConvNets is to characterize complex boundaries of thousands of classes in a\nhigh-dimensional space, it is critical to learn higher-order representations\nfor enhancing non-linear modeling capability. Recently, Global Second-order\nPooling (GSoP), plugged at the end of networks, has attracted increasing\nattentions, achieving much better performance than classical, first-order\nnetworks in a variety of vision tasks. However, how to effectively introduce\nhigher-order representation in earlier layers for improving non-linear\ncapability of ConvNets is still an open problem. In this paper, we propose a\nnovel network model introducing GSoP across from lower to higher layers for\nexploiting holistic image information throughout a network. Given an input 3D\ntensor outputted by some previous convolutional layer, we perform GSoP to\nobtain a covariance matrix which, after nonlinear transformation, is used for\ntensor scaling along channel dimension. Similarly, we can perform GSoP along\nspatial dimension for tensor scaling as well. In this way, we can make full use\nof the second-order statistics of the holistic image throughout a network. The\nproposed networks are thoroughly evaluated on large-scale ImageNet-1K, and\nexperiments have shown that they outperformed non-trivially the counterparts\nwhile achieving state-of-the-art results.","url_abs":"http://arxiv.org/abs/1811.12006v2","url_pdf":"http://arxiv.org/pdf/1811.12006v2.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":"global-second-order-pooling-convolutional","repo_url":"https://github.com/ZilinGao/Global-Second-order-Pooling-Convolutional-Networks","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"object-recognition","task_name":"Object Recognition"}],"methods":[{"method_slug":"gsop-net","method_name":"GSoP-Net"}],"datasets_introduced":[],"methods_introduced":[{"slug":"gsop-net","name":"GSoP-Net","full_name":"Global second-order pooling convolutional networks"}],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.12006","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1811.12006"}},"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/ZilinGao/Global-Second-order-Pooling-Convolutional-Networks","reach":null}],"summary":{"ran_draft_wrong":1,"ran_fixture":1,"unverified":1},"by_repo_kind":{"official":{"samples":2,"ran":1,"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":"fac5364e2f53c6db","entry":"conv3x3","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"fac5364e2f53c6db"}},{"code_sha256_prefix":"7bd6be9c711eba58","entry":"cov_feature","repo":"ZilinGao/Global-Second-order-Pooling-Convolutional-Networks","repo_kind":"official","path":"torchvision/models/resnet.py","file_url":"https://github.com/ZilinGao/Global-Second-order-Pooling-Convolutional-Networks/blob/HEAD/torchvision/models/resnet.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7bd6be9c711eba58"}},{"code_sha256_prefix":"f262b4e8c2114bd0","entry":"resnet50","repo":"ZilinGao/Global-Second-order-Pooling-Convolutional-Networks","repo_kind":"official","path":"torchvision/models/resnet.py","file_url":"https://github.com/ZilinGao/Global-Second-order-Pooling-Convolutional-Networks/blob/HEAD/torchvision/models/resnet.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f262b4e8c2114bd0"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}