{"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/lending-orientation-to-neural-networks-for","title":"Lending Orientation to Neural Networks for Cross-view Geo-localization","arxiv_id":"1903.12351","date":"2019-03-29","proceeding":"CVPR 2019 6","authors":["Liu Liu","Hongdong Li"],"abstract":"This paper studies image-based geo-localization (IBL) problem using\nground-to-aerial cross-view matching. The goal is to predict the spatial\nlocation of a ground-level query image by matching it to a large geotagged\naerial image database (e.g., satellite imagery). This is a challenging task due\nto the drastic differences in their viewpoints and visual appearances. Existing\ndeep learning methods for this problem have been focused on maximizing feature\nsimilarity between spatially close-by image pairs, while minimizing other\nimages pairs which are far apart. They do so by deep feature embedding based on\nvisual appearance in those ground-and-aerial images. However, in everyday life,\nhumans commonly use {\\em orientation} information as an important cue for the\ntask of spatial localization. Inspired by this insight, this paper proposes a\nnovel method which endows deep neural networks with the `commonsense' of\norientation. Given a ground-level spherical panoramic image as query input (and\na large georeferenced satellite image database), we design a Siamese network\nwhich explicitly encodes the orientation (i.e., spherical directions) of each\npixel of the images. Our method significantly boosts the discriminative power\nof the learned deep features, leading to a much higher recall and precision\noutperforming all previous methods. Our network is also more compact using only\n1/5th number of parameters than a previously best-performing network. To\nevaluate the generalization of our method, we also created a large-scale\ncross-view localization benchmark containing 100K geotagged ground-aerial pairs\ncovering a city. Our codes and datasets are available at\n\\url{https://github.com/Liumouliu/OriCNN}.","url_abs":"http://arxiv.org/abs/1903.12351v1","url_pdf":"http://arxiv.org/pdf/1903.12351v1.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":"lending-orientation-to-neural-networks-for","repo_url":"https://github.com/Liumouliu/OriCNN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"geo-localization","task_name":"geo-localization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1903.12351","atlas_url":"https://app.syntology.ai/?focus=1903.12351","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1903.12351"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Liumouliu/OriCNN","reach":null}],"summary":{"ran_honours":1},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"e6f037abe00e4e38","entry":"validate","repo":"Liumouliu/OriCNN","repo_kind":"official","path":"CVPR2019_codes/OriNet_CVACT/train_deep6_scratch_m1_1_concat3conv_rgb_ori_gem_ACT.py","file_url":"https://github.com/Liumouliu/OriCNN/blob/HEAD/CVPR2019_codes/OriNet_CVACT/train_deep6_scratch_m1_1_concat3conv_rgb_ori_gem_ACT.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e6f037abe00e4e38"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}