{"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-a-rotation-invariant-detector-with","title":"Learning a Rotation Invariant Detector with Rotatable Bounding Box","arxiv_id":"1711.09405","date":"2017-11-26","proceeding":null,"authors":["Lei Liu","Zongxu Pan","Bin Lei"],"abstract":"Detection of arbitrarily rotated objects is a challenging task due to the\ndifficulties of locating the multi-angle objects and separating them\neffectively from the background. The existing methods are not robust to angle\nvaries of the objects because of the use of traditional bounding box, which is\na rotation variant structure for locating rotated objects. In this article, a\nnew detection method is proposed which applies the newly defined rotatable\nbounding box (RBox). The proposed detector (DRBox) can effectively handle the\nsituation where the orientation angles of the objects are arbitrary. The\ntraining of DRBox forces the detection networks to learn the correct\norientation angle of the objects, so that the rotation invariant property can\nbe achieved. DRBox is tested to detect vehicles, ships and airplanes on\nsatellite images, compared with Faster R-CNN and SSD, which are chosen as the\nbenchmark of the traditional bounding box based methods. The results shows that\nDRBox performs much better than traditional bounding box based methods do on\nthe given tasks, and is more robust against rotation of input image and target\nobjects. Besides, results show that DRBox correctly outputs the orientation\nangles of the objects, which is very useful for locating multi-angle objects\nefficiently. The code and models are available at\nhttps://github.com/liulei01/DRBox.","url_abs":"http://arxiv.org/abs/1711.09405v1","url_pdf":"http://arxiv.org/pdf/1711.09405v1.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":"learning-a-rotation-invariant-detector-with","repo_url":"https://github.com/liulei01/DRBox","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"learning-a-rotation-invariant-detector-with","repo_url":"https://github.com/toshi-k/kaggle-3d-object-detection-for-autonomous-vehicles","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-a-rotation-invariant-detector-with","repo_url":"https://github.com/toshi-k/kaggle-airbus-ship-detection-challenge","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-a-rotation-invariant-detector-with","repo_url":"https://github.com/witwitchayakarn/DRBox","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"faster-r-cnn","method_name":"Faster R-CNN"},{"method_slug":"non-maximum-suppression","method_name":"Non Maximum Suppression"},{"method_slug":"rpn","method_name":"RPN"},{"method_slug":"roipool","method_name":"RoIPool"},{"method_slug":"ssd","method_name":"SSD"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1711.09405","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}