{"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-learning-for-large-scale-traffic-sign","title":"Deep Learning for Large-Scale Traffic-Sign Detection and Recognition","arxiv_id":"1904.00649","date":"2019-04-01","proceeding":null,"authors":["Domen Tabernik","Danijel Skočaj"],"abstract":"Automatic detection and recognition of traffic signs plays a crucial role in\nmanagement of the traffic-sign inventory. It provides accurate and timely way\nto manage traffic-sign inventory with a minimal human effort. In the computer\nvision community the recognition and detection of traffic signs is a\nwell-researched problem. A vast majority of existing approaches perform well on\ntraffic signs needed for advanced drivers-assistance and autonomous systems.\nHowever, this represents a relatively small number of all traffic signs (around\n50 categories out of several hundred) and performance on the remaining set of\ntraffic signs, which are required to eliminate the manual labor in traffic-sign\ninventory management, remains an open question. In this paper, we address the\nissue of detecting and recognizing a large number of traffic-sign categories\nsuitable for automating traffic-sign inventory management. We adopt a\nconvolutional neural network (CNN) approach, the Mask R-CNN, to address the\nfull pipeline of detection and recognition with automatic end-to-end learning.\nWe propose several improvements that are evaluated on the detection of traffic\nsigns and result in an improved overall performance. This approach is applied\nto detection of 200 traffic-sign categories represented in our novel dataset.\nResults are reported on highly challenging traffic-sign categories that have\nnot yet been considered in previous works. We provide comprehensive analysis of\nthe deep learning method for the detection of traffic signs with large\nintra-category appearance variation and show below 3% error rates with the\nproposed approach, which is sufficient for deployment in practical applications\nof traffic-sign inventory management.","url_abs":"http://arxiv.org/abs/1904.00649v1","url_pdf":"http://arxiv.org/pdf/1904.00649v1.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-learning-for-large-scale-traffic-sign","repo_url":"https://github.com/skokec/detectron-traffic-signs","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"caffe2","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"management","task_name":"Management"},{"task_slug":"open-question","task_name":"Open-Ended Question Answering"},{"task_slug":"traffic-sign-detection","task_name":"Traffic Sign Detection"},{"task_slug":"traffic-sign-recognition","task_name":"Traffic Sign Recognition"}],"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":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"mask-r-cnn","method_name":"Mask R-CNN"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"rpn","method_name":"RPN"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"roi-align","method_name":"RoIAlign"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/traffic-sign-recognition-on-dfg-traffic-sign-1","task":"Traffic Sign Recognition","dataset":"DFG traffic-sign dataset","model":"Mask R-CNN with adaptations for traffic sings and augmentations (ResNet50)","rank_in_archive_order":1,"of":5,"metrics":{"mAP @0.5:0.95":"84.4","mAP@0.50":"95.5"},"uses_additional_data":true},{"leaderboard":"/sota/traffic-sign-recognition-on-dfg-traffic-sign-1","task":"Traffic Sign Recognition","dataset":"DFG traffic-sign dataset","model":"Mask R-CNN with adaptations for traffic sings  (ResNet50)","rank_in_archive_order":2,"of":5,"metrics":{"mAP@0.50":"95.2"},"uses_additional_data":false},{"leaderboard":"/sota/traffic-sign-recognition-on-dfg-traffic-sign-1","task":"Traffic Sign Recognition","dataset":"DFG traffic-sign dataset","model":"Mask R-CNN (ResNet50)","rank_in_archive_order":3,"of":5,"metrics":{"mAP @0.5:0.95":"82.3","mAP@0.50":"93.0"},"uses_additional_data":false},{"leaderboard":"/sota/traffic-sign-recognition-on-dfg-traffic-sign-1","task":"Traffic Sign Recognition","dataset":"DFG traffic-sign dataset","model":"Faster R-CNN","rank_in_archive_order":4,"of":5,"metrics":{"mAP @0.5:0.95":"80.4","mAP@0.50":"92.4"},"uses_additional_data":false},{"leaderboard":"/sota/traffic-sign-recognition-on-dfg-traffic-sign-1","task":"Traffic Sign Recognition","dataset":"DFG traffic-sign dataset","model":"Mask R-CNN with adaptations for traffic sings (ResNet50)","rank_in_archive_order":5,"of":5,"metrics":{"mAP @0.5:0.95":"82.0"},"uses_additional_data":false},{"leaderboard":"/sota/traffic-sign-recognition-on-swedish-traffic-2","task":"Traffic Sign Recognition","dataset":"Swedish traffic-sign dataset (STSD)","model":"Mask R-CNN with adaptations for traffic sings (ResNet50)","rank_in_archive_order":1,"of":2,"metrics":{"mAP@0.50":"95.2"},"uses_additional_data":false},{"leaderboard":"/sota/traffic-sign-recognition-on-swedish-traffic-2","task":"Traffic Sign Recognition","dataset":"Swedish traffic-sign dataset (STSD)","model":"Faster R-CNN","rank_in_archive_order":2,"of":2,"metrics":{"mAP@0.50":"94.3"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1904.00649","atlas_url":"https://app.syntology.ai/?focus=1904.00649","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1904.00649"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/skokec/detectron-traffic-signs","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":2},"by_repo_kind":{"official":{"samples":2,"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":"06e62194f17cd06a","entry":"parse_rec","repo":"skokec/detectron-traffic-signs","repo_kind":"official","path":"lib/datasets/voc_eval.py","file_url":"https://github.com/skokec/detectron-traffic-signs/blob/HEAD/lib/datasets/voc_eval.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"06e62194f17cd06a"}},{"code_sha256_prefix":"611c47c44c424dba","entry":"voc_ap","repo":"skokec/detectron-traffic-signs","repo_kind":"official","path":"lib/datasets/voc_eval.py","file_url":"https://github.com/skokec/detectron-traffic-signs/blob/HEAD/lib/datasets/voc_eval.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"611c47c44c424dba"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}