{"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/accurate-single-stage-detector-using","title":"Accurate Single Stage Detector Using Recurrent Rolling Convolution","arxiv_id":"1704.05776","date":"2017-04-19","proceeding":"CVPR 2017 7","authors":["Jimmy Ren","Xiaohao Chen","Jianbo Liu","Wenxiu Sun","Jiahao Pang","Qiong Yan","Yu-Wing Tai","Li Xu"],"abstract":"Most of the recent successful methods in accurate object detection and\nlocalization used some variants of R-CNN style two stage Convolutional Neural\nNetworks (CNN) where plausible regions were proposed in the first stage then\nfollowed by a second stage for decision refinement. Despite the simplicity of\ntraining and the efficiency in deployment, the single stage detection methods\nhave not been as competitive when evaluated in benchmarks consider mAP for high\nIoU thresholds. In this paper, we proposed a novel single stage end-to-end\ntrainable object detection network to overcome this limitation. We achieved\nthis by introducing Recurrent Rolling Convolution (RRC) architecture over\nmulti-scale feature maps to construct object classifiers and bounding box\nregressors which are \"deep in context\". We evaluated our method in the\nchallenging KITTI dataset which measures methods under IoU threshold of 0.7. We\nshowed that with RRC, a single reduced VGG-16 based model already significantly\noutperformed all the previously published results. At the time this paper was\nwritten our models ranked the first in KITTI car detection (the hard level),\nthe first in cyclist detection and the second in pedestrian detection. These\nresults were not reached by the previous single stage methods. The code is\npublicly available.","url_abs":"http://arxiv.org/abs/1704.05776v1","url_pdf":"http://arxiv.org/pdf/1704.05776v1.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":"accurate-single-stage-detector-using","repo_url":"https://github.com/xiaohaoChen/rrc_detection","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"accurate-single-stage-detector-using","repo_url":"https://github.com/JunaidCS032/MOTBeyondPixels","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"3d-object-detection","task_name":"3D Object Detection"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"pedestrian-detection","task_name":"Pedestrian Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"faster-r-cnn","method_name":"Faster R-CNN"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"rpn","method_name":"RPN"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"roipool","method_name":"RoIPool"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.05776","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}