{"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/selective-kernel-networks","title":"Selective Kernel Networks","arxiv_id":"1903.06586","date":"2019-03-15","proceeding":"CVPR 2019 6","authors":["Xiang Li","Wenhai Wang","Xiaolin Hu","Jian Yang"],"abstract":"In standard Convolutional Neural Networks (CNNs), the receptive fields of\nartificial neurons in each layer are designed to share the same size. It is\nwell-known in the neuroscience community that the receptive field size of\nvisual cortical neurons are modulated by the stimulus, which has been rarely\nconsidered in constructing CNNs. We propose a dynamic selection mechanism in\nCNNs that allows each neuron to adaptively adjust its receptive field size\nbased on multiple scales of input information. A building block called\nSelective Kernel (SK) unit is designed, in which multiple branches with\ndifferent kernel sizes are fused using softmax attention that is guided by the\ninformation in these branches. Different attentions on these branches yield\ndifferent sizes of the effective receptive fields of neurons in the fusion\nlayer. Multiple SK units are stacked to a deep network termed Selective Kernel\nNetworks (SKNets). On the ImageNet and CIFAR benchmarks, we empirically show\nthat SKNet outperforms the existing state-of-the-art architectures with lower\nmodel complexity. Detailed analyses show that the neurons in SKNet can capture\ntarget objects with different scales, which verifies the capability of neurons\nfor adaptively adjusting their receptive field sizes according to the input.\nThe code and models are available at https://github.com/implus/SKNet.","url_abs":"http://arxiv.org/abs/1903.06586v2","url_pdf":"http://arxiv.org/pdf/1903.06586v2.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":"selective-kernel-networks","repo_url":"https://github.com/implus/SKNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"gone","observed_at":"2026-09-18","how":"tree_404+repo_404"}},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/ChriXiang/Selective_Kernel_Keras","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/Mind23-2/MindCode-79","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/YeongHyeon/SKNet-TF2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/developer0hye/SKNet-PyTorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/frgfm/Holocron","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/implus/PytorchInsight","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/osmr/imgclsmob","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mxnet","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/rwightman/pytorch-image-models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/whai362/PytorchInsight","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/xmu-xiaoma666/External-Attention-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/ywj95/Example-based-Bokeh-Effect-Challenge","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/2023-MindSpore-1/ms-code-47","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/2023-MindSpore-4/Code11/tree/main/sknet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/2023-MindSpore-4/Code7/tree/main/sknet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/MS-Mind/MS-Code-02/tree/main/configs/sknet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/Mind23-2/MindCode-88/tree/main/sknet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/mindspore-ai/models/tree/master/research/cv/sknet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/mindspore-courses/External-Attention-MindSpore/blob/main/model/attention/SKAttention.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"selective-kernel-networks","repo_url":"https://github.com/mindspore-ecosystem/mindcv/blob/main/mindcv/models/sknet.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"}],"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":"convolution","method_name":"Convolution"},{"method_slug":"dilated-convolution","method_name":"Dilated Convolution"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"grouped-convolution","method_name":"Grouped Convolution"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"resnext","method_name":"ResNeXt"},{"method_slug":"resnext-block","method_name":"ResNeXt Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"selective-kernel","method_name":"Selective Kernel"},{"method_slug":"selective-kernel-convolution","method_name":"Selective Kernel Convolution"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[{"slug":"selective-kernel","name":"Selective Kernel","full_name":"Selective Kernel"},{"slug":"selective-kernel-convolution","name":"Selective Kernel Convolution","full_name":"Selective Kernel Convolution"}],"results":[{"leaderboard":"/sota/image-classification-on-cifar-10","task":"Image Classification","dataset":"CIFAR-10","model":"SKNet-29 (ResNeXt-29, 16×32d)","rank_in_archive_order":107,"of":265,"metrics":{"Percentage correct":"96.53"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-cifar-100","task":"Image Classification","dataset":"CIFAR-100","model":"SKNet-29 (ResNeXt-29, 16×32d)","rank_in_archive_order":100,"of":211,"metrics":{"Percentage correct":"82.67"},"uses_additional_data":true},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"SKNet-101","rank_in_archive_order":733,"of":1060,"metrics":{"GFLOPs":"8.46","Number of params":"48.9M","Top 1 Accuracy":"79.81%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.06586","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1903.06586"}},"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/rwightman/pytorch-image-models","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/implus/PytorchInsight","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2023-MindSpore-4/Code11/tree/main/sknet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/frgfm/Holocron","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/implus/SKNet","reach":{"status":"gone","observed_at":"2026-09-18","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/whai362/PytorchInsight","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Mind23-2/MindCode-79","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mindspore-courses/External-Attention-MindSpore/blob/main/model/attention/SKAttention.py","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MS-Mind/MS-Code-02/tree/main/configs/sknet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/osmr/imgclsmob","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/YeongHyeon/SKNet-TF2","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2023-MindSpore-4/Code7/tree/main/sknet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ywj95/Example-based-Bokeh-Effect-Challenge","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/xmu-xiaoma666/External-Attention-pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2023-MindSpore-1/ms-code-47","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mindspore-ai/models/tree/master/research/cv/sknet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mindspore-ecosystem/mindcv/blob/main/mindcv/models/sknet.py","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ChriXiang/Selective_Kernel_Keras","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Mind23-2/MindCode-88/tree/main/sknet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/developer0hye/SKNet-PyTorch","reach":null}],"summary":{"ran_draft_wrong":3},"by_repo_kind":{"listed":{"samples":3,"ran":3,"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":3,"samples":[{"code_sha256_prefix":"1463a3cc54f82dba","entry":"SKNet101","repo":"developer0hye/SKNet-PyTorch","repo_kind":"listed","path":"sknet.py","file_url":"https://github.com/developer0hye/SKNet-PyTorch/blob/HEAD/sknet.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1463a3cc54f82dba"}},{"code_sha256_prefix":"c64972575dab1437","entry":"SKNet26","repo":"developer0hye/SKNet-PyTorch","repo_kind":"listed","path":"sknet.py","file_url":"https://github.com/developer0hye/SKNet-PyTorch/blob/HEAD/sknet.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c64972575dab1437"}},{"code_sha256_prefix":"fe7329baca7b42a2","entry":"SKNet50","repo":"developer0hye/SKNet-PyTorch","repo_kind":"listed","path":"sknet.py","file_url":"https://github.com/developer0hye/SKNet-PyTorch/blob/HEAD/sknet.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fe7329baca7b42a2"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}