{"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/dfanet-deep-feature-aggregation-for-real-time","title":"DFANet: Deep Feature Aggregation for Real-Time Semantic Segmentation","arxiv_id":"1904.02216","date":"2019-04-03","proceeding":"CVPR 2019 6","authors":["Hanchao Li","Pengfei Xiong","Haoqiang Fan","Jian Sun"],"abstract":"This paper introduces an extremely efficient CNN architecture named DFANet\nfor semantic segmentation under resource constraints. Our proposed network\nstarts from a single lightweight backbone and aggregates discriminative\nfeatures through sub-network and sub-stage cascade respectively. Based on the\nmulti-scale feature propagation, DFANet substantially reduces the number of\nparameters, but still obtains sufficient receptive field and enhances the model\nlearning ability, which strikes a balance between the speed and segmentation\nperformance. Experiments on Cityscapes and CamVid datasets demonstrate the\nsuperior performance of DFANet with 8$\\times$ less FLOPs and 2$\\times$ faster\nthan the existing state-of-the-art real-time semantic segmentation methods\nwhile providing comparable accuracy. Specifically, it achieves 70.3\\% Mean IOU\non the Cityscapes test dataset with only 1.7 GFLOPs and a speed of 160 FPS on\none NVIDIA Titan X card, and 71.3\\% Mean IOU with 3.4 GFLOPs while inferring on\na higher resolution image.","url_abs":"http://arxiv.org/abs/1904.02216v1","url_pdf":"http://arxiv.org/pdf/1904.02216v1.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":"dfanet-deep-feature-aggregation-for-real-time","repo_url":"https://github.com/j-a-lin/DFANet_PyTorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"dfanet-deep-feature-aggregation-for-real-time","repo_url":"https://github.com/huaifeng1993/DFANet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"real-time-semantic-segmentation","task_name":"Real-Time Semantic Segmentation"},{"task_slug":"smac-1","task_name":"SMAC+"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/smac-on-smac-def-infantry-parallel","task":"SMAC+","dataset":"Def_Infantry_parallel","model":"DIQL","rank_in_archive_order":7,"of":10,"metrics":{"Median Win Rate":"45.0"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-camvid","task":"Semantic Segmentation","dataset":"CamVid","model":"DFANet A","rank_in_archive_order":15,"of":21,"metrics":{"Mean IoU":"64.7%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-cityscapes","task":"Semantic Segmentation","dataset":"Cityscapes test","model":"DFANet A","rank_in_archive_order":75,"of":105,"metrics":{"Mean IoU (class)":"71.3%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.02216","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1904.02216"}},"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/huaifeng1993/DFANet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/j-a-lin/DFANet_PyTorch","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":1},"by_repo_kind":{"listed":{"samples":1,"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":"dba26c07abbe7567","entry":"load_backbone","repo":"huaifeng1993/DFANet","repo_kind":"listed","path":"model/dfanet.py","file_url":"https://github.com/huaifeng1993/DFANet/blob/HEAD/model/dfanet.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":"dba26c07abbe7567"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}