{"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/learning-multilayer-channel-features-for","title":"Learning Multilayer Channel Features for Pedestrian Detection","arxiv_id":"1603.00124","date":"2016-03-01","proceeding":null,"authors":["Jiale Cao","Yanwei Pang","Xuelong. Li"],"abstract":"Pedestrian detection based on the combination of Convolutional Neural Network\n(i.e., CNN) and traditional handcrafted features (i.e., HOG+LUV) has achieved\ngreat success. Generally, HOG+LUV are used to generate the candidate proposals\nand then CNN classifies these proposals. Despite its success, there is still\nroom for improvement. For example, CNN classifies these proposals by the\nfull-connected layer features while proposal scores and the features in the\ninner-layers of CNN are ignored. In this paper, we propose a unifying framework\ncalled Multilayer Channel Features (MCF) to overcome the drawback. It firstly\nintegrates HOG+LUV with each layer of CNN into a multi-layer image channels.\nBased on the multi-layer image channels, a multi-stage cascade AdaBoost is then\nlearned. The weak classifiers in each stage of the multi-stage cascade is\nlearned from the image channels of corresponding layer. With more abundant\nfeatures, MCF achieves the state-of-the-art on Caltech pedestrian dataset\n(i.e., 10.40% miss rate). Using new and accurate annotations, MCF achieves\n7.98% miss rate. As many non-pedestrian detection windows can be quickly\nrejected by the first few stages, it accelerates detection speed by 1.43 times.\nBy eliminating the highly overlapped detection windows with lower scores after\nthe first stage, it's 4.07 times faster with negligible performance loss.","url_abs":"http://arxiv.org/abs/1603.00124v1","url_pdf":"http://arxiv.org/pdf/1603.00124v1.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":[],"tasks":[{"task_slug":"pedestrian-detection","task_name":"Pedestrian Detection"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/pedestrian-detection-on-caltech","task":"Pedestrian Detection","dataset":"Caltech","model":"MCF","rank_in_archive_order":25,"of":33,"metrics":{"Reasonable Miss Rate":"10.40"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}