{"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/geolocation-estimation-of-photos-using-a","title":"Geolocation Estimation of Photos using a Hierarchical Model and Scene Classification","arxiv_id":null,"date":"2018-09-01","proceeding":"ECCV 2018 9","authors":["Eric Muller-Budack","Kader Pustu-Iren","Ralph Ewerth"],"abstract":"While the successful estimation of a photo's geolocation enables a number of interesting applications, it is also a very challenging task. Due to the complexity of the problem, most existing approaches are restricted to specific areas, imagery, or worldwide landmarks. Only a few proposals predict GPS coordinates without any limitations. In this paper, we introduce several deep learning methods, which pursue the latter approach and treat geolocalization as a classification problem where the earth is subdivided into geographical cells. We propose to exploit hierarchical knowledge of multiple partitionings and additionally extract and take the photo's scene content into account, i.e., indoor, natural, or urban setting etc. As a result, contextual information at different spatial resolutions as well as more specific features for various environmental settings are incorporated in the learning process of the convolutional neural network. Experimental results on two benchmarks demonstrate the effectiveness of our approach outperforming the state of the art while using a significant lower number of training images and without relying on retrieval methods that require an appropriate reference dataset.","url_abs":"http://openaccess.thecvf.com/content_ECCV_2018/html/Eric_Muller-Budack_Geolocation_Estimation_of_ECCV_2018_paper.html","url_pdf":"http://openaccess.thecvf.com/content_ECCV_2018/papers/Eric_Muller-Budack_Geolocation_Estimation_of_ECCV_2018_paper.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":"geolocation-estimation-of-photos-using-a","repo_url":"https://github.com/TIBHannover/GeoEstimation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"photo-geolocation-estimation","task_name":"Photo geolocation estimation"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"scene-classification","task_name":"Scene Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/photo-geolocation-estimation-on-gws15k","task":"Photo geolocation estimation","dataset":"GWS15k","model":"ISNs (M, f*, S3)","rank_in_archive_order":5,"of":5,"metrics":{"City level (25 km)":"0.6","Continent level (2500 km)":"38.5","Country level (750 km)":"15.5","Region level (200 km)":"4.2","Street level (1 km)":"0.05"},"uses_additional_data":false},{"leaderboard":"/sota/photo-geolocation-estimation-on-im2gps","task":"Photo geolocation estimation","dataset":"Im2GPS","model":"ISNs (M, f*, S3)","rank_in_archive_order":2,"of":11,"metrics":{"City level (25 km)":"43.0","Continent level (2500 km)":"80.2","Country level (750 km)":"66.7","Reference images":"0","Region level (200 km)":"51.9","Street level (1 km)":"16.9","Training images":"4.7M"},"uses_additional_data":false},{"leaderboard":"/sota/photo-geolocation-estimation-on-im2gps","task":"Photo geolocation estimation","dataset":"Im2GPS","model":"base (M, f*)","rank_in_archive_order":4,"of":11,"metrics":{"City level (25 km)":"40.9","Continent level (2500 km)":"78.5","Country level (750 km)":"65.4","Reference images":"0","Region level (200 km)":"51.5","Street level (1 km)":"15.2","Training images":"4.7M"},"uses_additional_data":false},{"leaderboard":"/sota/photo-geolocation-estimation-on-im2gps","task":"Photo geolocation estimation","dataset":"Im2GPS","model":"base (L, m)","rank_in_archive_order":6,"of":11,"metrics":{"City level (25 km)":"35.0","Continent level (2500 km)":"79.7","Country level (750 km)":"64.1","Reference images":"0","Region level (200 km)":"49.8","Street level (1 km)":"13.5","Training images":"4.7M"},"uses_additional_data":false},{"leaderboard":"/sota/photo-geolocation-estimation-on-im2gps3k","task":"Photo geolocation estimation","dataset":"Im2GPS3k","model":"ISNs (M, f*, S3)","rank_in_archive_order":7,"of":14,"metrics":{"City level (25 km)":"28.0","Continent level (2500 km)":"66.0","Country level (750 km)":"49.7","Region level (200 km)":"36.6","Street level (1 km)":"10.5","Training Images":"4.7M"},"uses_additional_data":false},{"leaderboard":"/sota/photo-geolocation-estimation-on-im2gps3k","task":"Photo geolocation estimation","dataset":"Im2GPS3k","model":"base (M, f*)","rank_in_archive_order":9,"of":14,"metrics":{"City level (25 km)":"27.0","Continent level (2500 km)":"66.0","Country level (750 km)":"49.2","Region level (200 km)":"35.6","Street level (1 km)":"9.7","Training Images":"4.7M"},"uses_additional_data":false},{"leaderboard":"/sota/photo-geolocation-estimation-on-im2gps3k","task":"Photo geolocation estimation","dataset":"Im2GPS3k","model":"base (L, m)","rank_in_archive_order":10,"of":14,"metrics":{"City level (25 km)":"24.9","Continent level (2500 km)":"65.8","Country level (750 km)":"48.8","Region level (200 km)":"34.0","Street level (1 km)":"8.3","Training Images":"4.7M"},"uses_additional_data":false},{"leaderboard":"/sota/photo-geolocation-estimation-on-yfcc26k","task":"Photo geolocation estimation","dataset":"YFCC26k","model":"ISNs (M, f*, S3)","rank_in_archive_order":5,"of":6,"metrics":{"City level (25 km)":"12.3","Continent level (2500 km)":"50.7","Country level (750 km)":" 31.9","Region level (200 km)":"19.0","Street level (1 km)":"5.3","Training Images":"4.7M"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}