{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T11:53:47Z","timestamp":1761738827833,"version":"build-2065373602"},"reference-count":44,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2025,10,26]],"date-time":"2025-10-26T00:00:00Z","timestamp":1761436800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61971356"],"award-info":[{"award-number":["61971356"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100017596","name":"Natural Science Basic Research Program of Shaanxi Province","doi-asserted-by":"crossref","award":["2024JC-DXWT-07","2024JC-YBQN-0719"],"award-info":[{"award-number":["2024JC-DXWT-07","2024JC-YBQN-0719"]}],"id":[{"id":"10.13039\/501100017596","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJGI"],"abstract":"<jats:p>Image-based geo-localization is a challenging task that aims to determine the geographic location of a ground-level query image captured by an Unmanned Ground Vehicle (UGV) by matching it to geo-tagged nadir-view (top-down) images from an Unmanned Aerial Vehicle (UAV) stored in a reference database. The challenge comes from the perspective inconsistency between matched objects. In this work, we propose a novel metric learning scheme for hard exemplar mining to improve the performance of cross-view geo-localization. Specifically, we introduce a Dynamic Online Cross-Batch (DOCB) hard exemplar mining scheme that solves the problem of the lack of hard exemplars in mini-batches in the middle and late stages of training, which leads to training stagnation. It mines cross-batch hard negative exemplars according to the current network state and reloads them into the network to make the gradient of negative exemplars participating in back-propagation. Since the feature representation of cross-batch negative examples adapts to the current network state, the triplet loss calculation becomes more accurate. Compared with methods only considering the gradient of anchors and positives, adding the gradient of negative exemplars helps us to obtain the correct gradient direction. Therefore, our DOCB scheme can better guide the network to learn valuable metric information. Moreover, we design a simple Siamese-like network called multi-scale feature aggregation (MSFA), which can generate multi-scale feature aggregation by learning and fusing multiple local spatial embeddings. The experimental results demonstrate that our DOCB scheme and MSFA network achieve an accuracy of 95.78% on the CVUSA dataset and 86.34% on the CVACT_val dataset, which outperforms those of other existing methods in the field.<\/jats:p>","DOI":"10.3390\/ijgi14110418","type":"journal-article","created":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T02:23:28Z","timestamp":1761704608000},"page":"418","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["DOCB: A Dynamic Online Cross-Batch Hard Exemplar Recall for Cross-View Geo-Localization"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-7392-9648","authenticated-orcid":false,"given":"Wenchao","family":"Fan","sequence":"first","affiliation":[{"name":"School of Computer Science, Northwestern Polytechnical University, Xi\u2019an 710072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xuetao","family":"Tian","sequence":"additional","affiliation":[{"name":"School of Computer Science, Northwestern Polytechnical University, Xi\u2019an 710072, China"},{"name":"Xi\u2019an ASN Technology Group Co., Ltd., Xi\u2019an 710065, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Long","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Computer Science, Northwestern Polytechnical University, Xi\u2019an 710072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7230-1476","authenticated-orcid":false,"given":"Xiuwei","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer Science, Northwestern Polytechnical University, Xi\u2019an 710072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fang","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science, Northwestern Polytechnical University, Xi\u2019an 710072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,10,26]]},"reference":[{"doi-asserted-by":"crossref","unstructured":"Shi, Y., Yu, X., Liu, L., Zhang, T., and Li, H. (2020, January 7\u201312). Optimal feature transport for cross-view image geo-localization. Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA.","key":"ref_1","DOI":"10.1609\/aaai.v34i07.6875"},{"doi-asserted-by":"crossref","unstructured":"Liu, L., and Li, H. (2019, January 15\u201320). Lending orientation to neural networks for cross-view geo-localization. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.","key":"ref_2","DOI":"10.1109\/CVPR.2019.00577"},{"key":"ref_3","first-page":"1","article-title":"Spatial-aware feature aggregation for image based cross-view geo-localization","volume":"32","author":"Shi","year":"2019","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_4","first-page":"29009","article-title":"Cross-view geo-localization with layer-to-layer transformer","volume":"34","author":"Yang","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"unstructured":"Wang, T., Fan, S., Liu, D., and Sun, C. (2022). Transformer-guided convolutional neural network for cross-view geolocalization. arXiv.","key":"ref_5"},{"doi-asserted-by":"crossref","unstructured":"Zhu, S., Shah, M., and Chen, C. (2022, January 18\u201324). Transgeo: Transformer is all you need for cross-view image geo-localization. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","key":"ref_6","DOI":"10.1109\/CVPR52688.2022.00123"},{"unstructured":"Cai, S., Guo, Y., Khan, S., Hu, J., and Wen, G. (November, January 27). Ground-to-aerial image geo-localization with a hard exemplar reweighting triplet loss. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Seoul, Republic of Korea.","key":"ref_7"},{"doi-asserted-by":"crossref","unstructured":"Zhu, S., Yang, T., and Chen, C. (2021, January 5\u20139). Revisiting street-to-aerial view image geo-localization and orientation estimation. Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision, Virtual.","key":"ref_8","DOI":"10.1109\/WACV48630.2021.00080"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"2094","DOI":"10.1109\/TIP.2022.3152046","article-title":"Soft exemplar highlighting for cross-view image-based geo-localization","volume":"31","author":"Guo","year":"2022","journal-title":"IEEE Trans. Image Process."},{"doi-asserted-by":"crossref","unstructured":"Wang, X., Zhang, H., Huang, W., and Scott, M.R. (2020, January 14\u201319). Cross-batch memory for embedding learning. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","key":"ref_10","DOI":"10.1109\/CVPR42600.2020.00642"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"3224","DOI":"10.1109\/TIP.2021.3137005","article-title":"Cross-batch hard example mining with pseudo large batch for id vs. spot face recognition","volume":"31","author":"Tan","year":"2022","journal-title":"IEEE Trans. Image Process."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1023\/B:VISI.0000029664.99615.94","article-title":"Distinctive image features from scale-invariant keypoints","volume":"60","author":"Lowe","year":"2004","journal-title":"Int. J. Comput. Vis."},{"doi-asserted-by":"crossref","unstructured":"Bay, H., Tuytelaars, T., and Van Gool, L. (2006, January 7\u201313). Surf: Speeded up robust features. Proceedings of the Computer Vision\u2013ECCV 2006: 9th European Conference on Computer Vision, Graz, Austria. Proceedings, Part I 9.","key":"ref_13","DOI":"10.1007\/11744023_32"},{"doi-asserted-by":"crossref","unstructured":"Sivic, J., and Zisserman, A. (2006). Video Google: Efficient visual search of videos. Toward Category-Level Object Recognition, Springer.","key":"ref_14","DOI":"10.1007\/11957959_7"},{"doi-asserted-by":"crossref","unstructured":"Vo, N., Jacobs, N., and Hays, J. (2017, January 22\u201329). Revisiting im2gps in the deep learning era. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","key":"ref_15","DOI":"10.1109\/ICCV.2017.286"},{"doi-asserted-by":"crossref","unstructured":"Torii, A., Arandjelovic, R., Sivic, J., Okutomi, M., and Pajdla, T. (2015, January 7\u201312). 24\/7 place recognition by view synthesis. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.","key":"ref_16","DOI":"10.1109\/CVPR.2015.7298790"},{"doi-asserted-by":"crossref","unstructured":"Workman, S., and Jacobs, N. (2015, January 7\u201312). On the location dependence of convolutional neural network features. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Boston, MA, USA.","key":"ref_17","DOI":"10.1109\/CVPRW.2015.7301385"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1452","DOI":"10.1109\/TPAMI.2017.2723009","article-title":"Places: A 10 million image database for scene recognition","volume":"40","author":"Zhou","year":"2017","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"doi-asserted-by":"crossref","unstructured":"Vo, N.N., and Hays, J. (2016, January 11\u201314). Localizing and orienting street views using overhead imagery. Proceedings of the European Conference on Computer Vision, Amsterdam, The Netherlands.","key":"ref_19","DOI":"10.1007\/978-3-319-46448-0_30"},{"doi-asserted-by":"crossref","unstructured":"Arandjelovic, R., Gronat, P., Torii, A., Pajdla, T., and Sivic, J. (2016, January 27\u201330). NetVLAD: CNN architecture for weakly supervised place recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.","key":"ref_20","DOI":"10.1109\/CVPR.2016.572"},{"doi-asserted-by":"crossref","unstructured":"Hu, S., Feng, M., Nguyen, R.M., and Lee, G.H. (2018, January 18\u201322). Cvm-net: Cross-view matching network for image-based ground-to-aerial geo-localization. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.","key":"ref_21","DOI":"10.1109\/CVPR.2018.00758"},{"doi-asserted-by":"crossref","unstructured":"Shi, Y., Yu, X., Campbell, D., and Li, H. (2020, January 14\u201319). Where am i looking at? Joint location and orientation estimation by cross-view matching. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","key":"ref_22","DOI":"10.1109\/CVPR42600.2020.00412"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"4709313","DOI":"10.1109\/TGRS.2022.3210195","article-title":"It\u2019s okay to be wrong: Cross-view geo-localization with step-adaptive iterative refinement","volume":"60","author":"Lu","year":"2022","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"4704716","DOI":"10.1109\/TGRS.2023.3307508","article-title":"Cross-view object geo-localization in a local region with satellite imagery","volume":"61","author":"Sun","year":"2023","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"doi-asserted-by":"crossref","unstructured":"Zheng, Z., Wei, Y., and Yang, Y. (2020, January 12\u201316). University-1652: A multi-view multi-source benchmark for drone-based geo-localization. Proceedings of the 28th ACM International Conference on Multimedia, Seattle, WA, USA.","key":"ref_25","DOI":"10.1145\/3394171.3413896"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"4376","DOI":"10.1109\/TCSVT.2021.3135013","article-title":"A transformer-based feature segmentation and region alignment method for UAV-view geo-localization","volume":"32","author":"Dai","year":"2021","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"doi-asserted-by":"crossref","unstructured":"Zhu, R., Yang, M., Yin, L., Wu, F., and Yang, Y. (2023). UAV\u2019s status is worth considering: A fusion representations matching method for geo-localization. Sensors, 23.","key":"ref_27","DOI":"10.3390\/s23020720"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"5622813","DOI":"10.1109\/TGRS.2024.3396912","article-title":"Direction-Guided Multi-Scale Feature Fusion Network for Geo-localization","volume":"62","author":"Lv","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"doi-asserted-by":"crossref","unstructured":"Zhu, S., Yang, T., and Chen, C. (2021, January 19\u201325). Vigor: Cross-view image geo-localization beyond one-to-one retrieval. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Virtual.","key":"ref_29","DOI":"10.1109\/CVPR46437.2021.00364"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1927","DOI":"10.1109\/TCSVT.2023.3293514","article-title":"DeHi: A decoupled hierarchical architecture for unaligned ground-to-aerial geo-localization","volume":"34","author":"Wang","year":"2023","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_31","first-page":"5909516","article-title":"Multibranch joint representation learning based on information fusion strategy for cross-view geo-localization","volume":"62","author":"Ge","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"503","DOI":"10.1109\/TCSVT.2020.2988034","article-title":"Hierarchical temporal modeling with mutual distance matching for video based person re-identification","volume":"31","author":"Li","year":"2020","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"2644","DOI":"10.1109\/TCSVT.2017.2711015","article-title":"Deep localized metric learning","volume":"28","author":"Duan","year":"2017","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"doi-asserted-by":"crossref","unstructured":"Shi, Y., and Li, H. (2022, January 18\u201324). Beyond cross-view image retrieval: Highly accurate vehicle localization using satellite image. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","key":"ref_34","DOI":"10.1109\/CVPR52688.2022.01650"},{"doi-asserted-by":"crossref","unstructured":"Workman, S., Souvenir, R., and Jacobs, N. (2015, January 7\u201313). Wide-area image geolocalization with aerial reference imagery. Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile.","key":"ref_35","DOI":"10.1109\/ICCV.2015.451"},{"doi-asserted-by":"crossref","unstructured":"Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and Fei-Fei, L. (2009, January 20\u201325). Imagenet: A large-scale hierarchical image database. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.","key":"ref_36","DOI":"10.1109\/CVPR.2009.5206848"},{"unstructured":"Loshchilov, I., and Hutter, F. (2017). Decoupled weight decay regularization. arXiv.","key":"ref_37"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"867","DOI":"10.1109\/TCSVT.2021.3061265","article-title":"Each part matters: Local patterns facilitate cross-view geo-localization","volume":"32","author":"Wang","year":"2021","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"ref_39","first-page":"8022905","article-title":"SSA-Net: Spatial scale attention network for image-based geo-localization","volume":"19","author":"Zhang","year":"2021","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"3780","DOI":"10.1109\/TIP.2022.3175601","article-title":"Joint representation learning and keypoint detection for cross-view geo-localization","volume":"31","author":"Lin","year":"2022","journal-title":"IEEE Trans. Image Process."},{"doi-asserted-by":"crossref","unstructured":"Toker, A., Zhou, Q., Maximov, M., and Leal-Taix\u00e9, L. (2021, January 19\u201325). Coming down to earth: Satellite-to-street view synthesis for geo-localization. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Virtual.","key":"ref_41","DOI":"10.1109\/CVPR46437.2021.00642"},{"doi-asserted-by":"crossref","unstructured":"Tian, Y., Deng, X., Zhu, Y., and Newsam, S. (2020, January 1\u20135). Cross-time and orientation-invariant overhead image geolocalization using deep local features. Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision, Snowmass Village, CO, USA.","key":"ref_42","DOI":"10.1109\/WACV45572.2020.9093403"},{"key":"ref_43","first-page":"4703818","article-title":"4scig: A four-branch framework to reduce the interference of sky area in cross-view image geo-localization","volume":"62","author":"Li","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"doi-asserted-by":"crossref","unstructured":"Li, X., Wang, W., Hu, X., and Yang, J. (2019, January 15\u201320). Selective kernel networks. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.","key":"ref_44","DOI":"10.1109\/CVPR.2019.00060"}],"container-title":["ISPRS International Journal of Geo-Information"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2220-9964\/14\/11\/418\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T05:28:57Z","timestamp":1761715737000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2220-9964\/14\/11\/418"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,26]]},"references-count":44,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2025,11]]}},"alternative-id":["ijgi14110418"],"URL":"https:\/\/doi.org\/10.3390\/ijgi14110418","relation":{},"ISSN":["2220-9964"],"issn-type":[{"type":"electronic","value":"2220-9964"}],"subject":[],"published":{"date-parts":[[2025,10,26]]}}}