{"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/deep-layer-aggregation","title":"Deep Layer Aggregation","arxiv_id":"1707.06484","date":"2017-07-20","proceeding":"CVPR 2018 6","authors":["Fisher Yu","Dequan Wang","Evan Shelhamer","Trevor Darrell"],"abstract":"Visual recognition requires rich representations that span levels from low to\nhigh, scales from small to large, and resolutions from fine to coarse. Even\nwith the depth of features in a convolutional network, a layer in isolation is\nnot enough: compounding and aggregating these representations improves\ninference of what and where. Architectural efforts are exploring many\ndimensions for network backbones, designing deeper or wider architectures, but\nhow to best aggregate layers and blocks across a network deserves further\nattention. Although skip connections have been incorporated to combine layers,\nthese connections have been \"shallow\" themselves, and only fuse by simple,\none-step operations. We augment standard architectures with deeper aggregation\nto better fuse information across layers. Our deep layer aggregation structures\niteratively and hierarchically merge the feature hierarchy to make networks\nwith better accuracy and fewer parameters. Experiments across architectures and\ntasks show that deep layer aggregation improves recognition and resolution\ncompared to existing branching and merging schemes. The code is at\nhttps://github.com/ucbdrive/dla.","url_abs":"http://arxiv.org/abs/1707.06484v3","url_pdf":"http://arxiv.org/pdf/1707.06484v3.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":"deep-layer-aggregation","repo_url":"https://github.com/ucbdrive/dla","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"deep-layer-aggregation","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":"deep-layer-aggregation","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":"deep-layer-aggregation","repo_url":"https://github.com/PaddlePaddle/PaddleClas","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"deep-layer-aggregation","repo_url":"https://github.com/PaddlePaddle/PaddleDetection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"}],"methods":[{"method_slug":"dla","method_name":"DLA"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.06484","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}