{"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/indoor-visual-positioning-aided-by-cnn-based","title":"Indoor Visual Positioning Aided by CNN-Based Image Retrieval: Training-Free, 3D Modeling-Free","arxiv_id":null,"date":"2018-06-15","proceeding":"journal 2018 6","authors":["Yujin Chen","Ruizhi Chen","Mengyun Liu","Aoran Xiao","Dewen Wu  and Shuheng Zhao"],"abstract":"Indoor localization is one of the fundamentals of location-based services (LBS) such as\r\nseamless indoor and outdoor navigation, location-based precision marketing, spatial cognition of\r\nrobotics, etc. Visual features take up a dominant part of the information that helps human and\r\nrobotics understand the environment, and many visual localization systems have been proposed.\r\nHowever, the problem of indoor visual localization has not been well settled due to the tough\r\ntrade-off of accuracy and cost. To better address this problem, a localization method based on image\r\nretrieval is proposed in this paper, which mainly consists of two parts. The first one is CNN-based\r\nimage retrieval phase, CNN features extracted by pre-trained deep convolutional neural networks\r\n(DCNNs) from images are utilized to compare the similarity, and the output of this part are the\r\nmatched images of the target image. The second one is pose estimation phase that computes accurate\r\nlocalization result. Owing to the robust CNN feature extractor, our scheme is applicable to complex\r\nindoor environments and easily transplanted to outdoor environments. The pose estimation scheme\r\nwas inspired by monocular visual odometer, therefore, only RGB images and poses of reference\r\nimages are needed for accurate image geo-localization. Furthermore, our method attempts to use\r\nlightweight datum to present the scene. To evaluate the performance, experiments are conducted,\r\nand the result demonstrates that our scheme can efficiently result in high location accuracy as well as\r\norientation estimation. Currently the positioning accuracy and usability enhanced compared with\r\nsimilar solutions. Furthermore, our idea has a good application foreground, because the algorithms\r\nof data acquisition and pose estimation are compatible with the current state of data expansion","url_abs":"https://www.mdpi.com/1424-8220/18/8/2692","url_pdf":"https://www.researchgate.net/publication/327060416_Indoor_Visual_Positioning_Aided_by_CNN-Based_Image_Retrieval_Training-Free_3D_Modeling-Free","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":"indoor-visual-positioning-aided-by-cnn-based","repo_url":"https://github.com/TerenceCYJ/VP-SLAM-SC-papers","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"indoor-localization","task_name":"Indoor Localization"},{"task_slug":"marketing","task_name":"Marketing"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"visual-localization","task_name":"Visual Localization"},{"task_slug":"geo-localization","task_name":"geo-localization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}