{"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/augmenting-convolutional-networks-with","title":"Augmenting Convolutional networks with attention-based aggregation","arxiv_id":"2112.13692","date":"2021-12-27","proceeding":null,"authors":["Hugo Touvron","Matthieu Cord","Alaaeldin El-Nouby","Piotr Bojanowski","Armand Joulin","Gabriel Synnaeve","Hervé Jégou"],"abstract":"We show how to augment any convolutional network with an attention-based global map to achieve non-local reasoning. We replace the final average pooling by an attention-based aggregation layer akin to a single transformer block, that weights how the patches are involved in the classification decision. We plug this learned aggregation layer with a simplistic patch-based convolutional network parametrized by 2 parameters (width and depth). In contrast with a pyramidal design, this architecture family maintains the input patch resolution across all the layers. It yields surprisingly competitive trade-offs between accuracy and complexity, in particular in terms of memory consumption, as shown by our experiments on various computer vision tasks: object classification, image segmentation and detection.","url_abs":"https://arxiv.org/abs/2112.13692v1","url_pdf":"https://arxiv.org/pdf/2112.13692v1.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":"augmenting-convolutional-networks-with","repo_url":"https://github.com/facebookresearch/deit","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"augmenting-convolutional-networks-with","repo_url":"https://github.com/DarshanDeshpande/jax-models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":null},{"paper_slug":"augmenting-convolutional-networks-with","repo_url":"https://github.com/dongkyuk/PatchConvNet-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"augmenting-convolutional-networks-with","repo_url":"https://github.com/keras-team/keras-io/blob/master/examples/vision/patch_convnet.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"augmenting-convolutional-networks-with","repo_url":"https://github.com/mindspore-courses/External-Attention-MindSpore/blob/main/model/backbone/PatchConvnet.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"ca","method_name":"Class Attention"},{"method_slug":"layerscale","method_name":"LayerScale"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PatchConvNet-L120-21k-384","rank_in_archive_order":108,"of":1060,"metrics":{"Number of params":"334.3M","Top 1 Accuracy":"87.1%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PatchConvNet-B60-21k-384","rank_in_archive_order":139,"of":1060,"metrics":{"Number of params":"99.4M","Top 1 Accuracy":"86.5%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PatchConvNet-S60-21k-512","rank_in_archive_order":230,"of":1060,"metrics":{"Number of params":"25.2M","Top 1 Accuracy":"85.4%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PatchConvNet-B120","rank_in_archive_order":360,"of":1060,"metrics":{"Number of params":"188.6M","Top 1 Accuracy":"84.1%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PatchConvNet-B60","rank_in_archive_order":423,"of":1060,"metrics":{"Number of params":"99.4M","Top 1 Accuracy":"83.5%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PatchConvNet-S120","rank_in_archive_order":458,"of":1060,"metrics":{"Number of params":"47.7M","Top 1 Accuracy":"83.2%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PatchConvNet-S60","rank_in_archive_order":574,"of":1060,"metrics":{"Number of params":"25.2M","Top 1 Accuracy":"82.1%"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-minival","task":"Object Detection","dataset":"COCO minival","model":"PatchConvNet-S120 (Mask R-CNN)","rank_in_archive_order":100,"of":220,"metrics":{"box AP":"47.0"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-minival","task":"Object Detection","dataset":"COCO minival","model":"PatchConvNet-S60 (Mask R-CNN)","rank_in_archive_order":106,"of":220,"metrics":{"box AP":"46.4"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"PatchConvNet-L120 (UperNet)","rank_in_archive_order":84,"of":235,"metrics":{"Validation mIoU":"52.9"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"PatchConvNet-B120\n(UperNet)","rank_in_archive_order":86,"of":235,"metrics":{"Validation mIoU":"52.8"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"PatchConvNet-B60 (UperNet)","rank_in_archive_order":102,"of":235,"metrics":{"Validation mIoU":"51.1"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k","task":"Semantic Segmentation","dataset":"ADE20K","model":"PatchConvNet-S60 (UperNet)","rank_in_archive_order":134,"of":235,"metrics":{"Validation mIoU":"49.3"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k-val","task":"Semantic Segmentation","dataset":"ADE20K val","model":"PatchConvNet-L120 (UperNet)","rank_in_archive_order":41,"of":95,"metrics":{"mIoU":"52.9"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k-val","task":"Semantic Segmentation","dataset":"ADE20K val","model":"PatchConvNet-B120 (UperNet)","rank_in_archive_order":43,"of":95,"metrics":{"mIoU":"52.8"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k-val","task":"Semantic Segmentation","dataset":"ADE20K val","model":"PatchConvNet-B60 (UperNet)","rank_in_archive_order":47,"of":95,"metrics":{"mIoU":"51.1"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-ade20k-val","task":"Semantic Segmentation","dataset":"ADE20K val","model":"PatchConvNet-S60 (UperNet)","rank_in_archive_order":59,"of":95,"metrics":{"mIoU":"49.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2112.13692","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2112.13692"}},"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/facebookresearch/deit","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/keras-team/keras-io/blob/master/examples/vision/patch_convnet.py","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dongkyuk/PatchConvNet-pytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/DarshanDeshpande/jax-models","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mindspore-courses/External-Attention-MindSpore/blob/main/model/backbone/PatchConvnet.py","reach":null}],"summary":{"ran_draft_wrong":1,"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1},"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":0,"samples":[{"code_sha256_prefix":"ba6aa5f07daca9cd","entry":"conv3x3","repo":"facebookresearch/deit","repo_kind":"official","path":"patchconvnet_models.py","file_url":"https://github.com/facebookresearch/deit/blob/HEAD/patchconvnet_models.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":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ba6aa5f07daca9cd"}},{"code_sha256_prefix":"af6a4caff0775c04","entry":"get_activation","repo":"dongkyuk/PatchConvNet-pytorch","repo_kind":"listed","path":"model/custom_blocks.py","file_url":"https://github.com/dongkyuk/PatchConvNet-pytorch/blob/HEAD/model/custom_blocks.py","link_basis":"harvester_set","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":"af6a4caff0775c04"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}