{"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/large-scale-visual-place-recognition-with-sub","title":"Large scale visual place recognition with sub-linear storage growth","arxiv_id":"1810.09660","date":"2018-10-23","proceeding":null,"authors":["Huu Le","Michael Milford"],"abstract":"Robotic and animal mapping systems share many of the same objectives and\nchallenges, but differ in one key aspect: where much of the research in robotic\nmapping has focused on solving the data association problem, the grid cell\nneurons underlying maps in the mammalian brain appear to intentionally break\ndata association by encoding many locations with a single grid cell neuron. One\npotential benefit of this intentional aliasing is both sub-linear map storage\nand computational requirements growth with environment size, which we\ndemonstrated in a previous proof-of-concept study that detected and encoded\nmutually complementary co-prime pattern frequencies in the visual map data. In\nthis research, we solve several of the key theoretical and practical\nlimitations of that prototype model and achieve significantly better sub-linear\nstorage growth, a factor reduction in storage requirements per map location,\nscalability to large datasets on standard compute equipment and improved\nrobustness to environments with visually challenging appearance change. These\nimprovements are achieved through several innovations including a flexible\nuser-driven choice mechanism for the periodic patterns underlying the new\nencoding method, a parallelized chunking technique that splits the map into\nsub-sections processed in parallel and a novel feature selection approach that\nselects only the image information most relevant to the encoded temporal\npatterns. We evaluate our techniques on two large benchmark datasets with the\ncomparison to the previous state-of-the-art system, as well as providing a\ndetailed analysis of system performance with respect to parameters such as\nrequired precision performance and the number of cyclic patterns encoded.","url_abs":"http://arxiv.org/abs/1810.09660v1","url_pdf":"http://arxiv.org/pdf/1810.09660v1.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":"large-scale-visual-place-recognition-with-sub","repo_url":"https://github.com/intellhave/SublinearEncoding","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"chunking","task_name":"Chunking"},{"task_slug":"visual-place-recognition","task_name":"Visual Place Recognition"},{"task_slug":"feature-selection","task_name":"feature selection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}