{"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/deepproposal-hunting-objects-by-cascading","title":"DeepProposal: Hunting Objects by Cascading Deep Convolutional Layers","arxiv_id":"1510.04445","date":"2015-10-15","proceeding":"ICCV 2015 12","authors":["Amir Ghodrati","Ali Diba","Marco Pedersoli","Tinne Tuytelaars","Luc van Gool"],"abstract":"In this paper we evaluate the quality of the activation layers of a\nconvolutional neural network (CNN) for the gen- eration of object proposals. We\ngenerate hypotheses in a sliding-window fashion over different activation\nlayers and show that the final convolutional layers can find the object of\ninterest with high recall but poor localization due to the coarseness of the\nfeature maps. Instead, the first layers of the network can better localize the\nobject of interest but with a reduced recall. Based on this observation we\ndesign a method for proposing object locations that is based on CNN features\nand that combines the best of both worlds. We build an inverse cascade that,\ngoing from the final to the initial convolutional layers of the CNN, selects\nthe most promising object locations and refines their boxes in a coarse-to-fine\nmanner. The method is efficient, because i) it uses the same features extracted\nfor detection, ii) it aggregates features using integral images, and iii) it\navoids a dense evaluation of the proposals due to the inverse coarse-to-fine\ncascade. The method is also accurate; it outperforms most of the previously\nproposed object proposals approaches and when plugged into a CNN-based detector\nproduces state-of-the- art detection performance.","url_abs":"http://arxiv.org/abs/1510.04445v1","url_pdf":"http://arxiv.org/pdf/1510.04445v1.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":"deepproposal-hunting-objects-by-cascading","repo_url":"https://github.com/aghodrati/deepproposal","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"object","task_name":"Object"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}