GEOCAPSNET: Ground to Aerial View Image Geo-Localization using Capsule Network

The task of cross-view image geo-localization aims to determine the geo-location (GPS coordinates) of a query ground-view image by matching it with the GPS-tagged aerial (satellite) images in a reference dataset. Due to the dramatic changes of viewpoint, matching the cross-view images is challenging. In this paper, we propose the GeoCapsNet based on the capsule network for ground-to-aerial image geo-localization. The network first extracts features from both ground and aerial images via standard convolution layers and the capsule layers further encode the features to model the spatial feature hierarchies and enhance the representation power. Moreover, we introduce a simple and effective weighted soft-margin triplet loss with online batch hard sample mining, which can greatly improve the image retrieval accuracy. Experimental results show that our GeoCapsNet significantly outperforms the state-of-the-art approaches on two benchmark datasets.

• Publication year

  2019

• Venue

  IEEE International Conference on Multimedia and Expo

• Publication date

  2019-04-12

• Fields of study

  Computer Science, Engineering, Environmental Science

• Identifiers
  DOI 10.1109/ICME.2019.00133arXiv 1904.06281
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• CVM-Net: Cross-View Matching Network for Image-Based Ground-to-Aerial Geo-Localization

  2018influential reference
• NetVLAD: CNN Architecture for Weakly Supervised Place Recognition

  2018cited by this paper
• Cross-View Image Matching for Geo-Localization in Urban Environments

  2017cited by this paper
• Beyond Triplet Loss: A Deep Quadruplet Network for Person Re-identification

  2017cited by this paper
• Dynamic Routing Between Capsules

  2017influential reference
• Predicting Ground-Level Scene Layout from Aerial Imagery

  2016influential reference
• Person Re-identification by Multi-Channel Parts-Based CNN with Improved Triplet Loss Function

  2016cited by this paper
• Robust Visual Place Recognition with Graph Kernels

  2016cited by this paper
• Gated Siamese Convolutional Neural Network Architecture for Human Re-identification

  2016cited by this paper
• Localizing and Orienting Street Views Using Overhead Imagery

  2016influential reference
• Wide-Area Image Geolocalization with Aerial Reference Imagery

  2015cited by this paper
• On the location dependence of convolutional neural network features

  2015cited by this paper
• Learning deep representations for ground-to-aerial geolocalization

  2015cited by this paper
• FaceNet: A unified embedding for face recognition and clustering

  2015cited by this paper
• Deep Residual Learning for Image Recognition

  2015influential reference
• Image Geo-Localization Based on MultipleNearest Neighbor Feature Matching UsingGeneralized Graphs

  2014cited by this paper
• Adam: A Method for Stochastic Optimization

  2014cited by this paper
• Accurate Geo-Registration by Ground-to-Aerial Image Matching

  2014cited by this paper
• Cross-View Image Geolocalization

  2013cited by this paper
• Ultra-wide Baseline Facade Matching for Geo-localization

  2012cited by this paper
• Visual localization by linear combination of image descriptors

  2011cited by this paper
• Accurate Image Localization Based on Google Maps Street View

  2010cited by this paper
• IM2GPS: estimating geographic information from a single image

  2008cited by this paper
• City-Scale Location Recognition

  2007cited by this paper
• Learning a similarity metric discriminatively, with application to face verification

  2005cited by this paper
• Image registration methods: a survey

  2003cited by this paper

• GazeViT: A gaze-guided hybrid attention vision transformer for cross-view matching of street-to-aerial images

  2025cites this paper
• Semantic Concept Perception Network With Interactive Prompting for Cross-View Image Geo-Localization

  2025cites this paper
• CIG2S: A Cross-View Image Geo-Localization Model Based on G2S Transform Suitable for Center-Misaligned Scenarios

  2025cites this paper
• VAGeo: View-specific Attention for Cross-View Object Geo-Localization

  2025cites this paper
• AMPLE: Automatic Progressive Learning for Orientation Unknown Ground-to-Aerial Geo-Localization

  2025cites this paper
• Learnable Query Aggregation with KV Routing for Cross-view Geo-localisation

  2025cites this paper
• Cross-View Image Geo-localization Based on Attention Weight Masks

  2025cites this paper
• HALO: Hierarchical Adaptive Feature Learning and Cross-View Interaction for Object-Level Geo-Localization

  2025cites this paper
• CCIGeo: Cross-View and Cross-Day-Night Image Geo-Localization Using Daytime Image Supervision

  2025cites this paper
• Cross-view geo-localization with panoramic street-view and VHR satellite imagery in decentrality settings

  2025cites this paper
• Offset Regression Enhanced Cross-View Feature Interaction for Ground-to-Aerial Geo-Localization

  2025cites this paper
• Patch Similarity Self-Knowledge Distillation for Cross-View Geo-Localization

  2024cites this paper
• TransFG: A Cross-View Geo-Localization of Satellite and UAVs Imagery Pipeline Using Transformer-Based Feature Aggregation and Gradient Guidance

  2024cites this paper
• GeoViewMatch: A Multi-Scale Feature-Matching Network for Cross-View Geo-Localization Using Swin-Transformer and Contrastive Learning

  2024cites this paper
• Ground–Satellite Coupling for Cross-View Geolocation Combined With Multiscale Fusion of Spatial Features

  2024cites this paper
• MT-GEO: A Multi-Scale Feature Extraction Network for Cross-View Geo-Localization Between Street-View and Remote Sensing Imagery

  2024cites this paper
• Cross-View Geo-Localization with Street-View and VHR Satellite Imagery in Decentrality Settings

  2024cites this paper
• Geometry-Guided Cross-View Diffusion for One-to-Many Cross-View Image Synthesis

  2024cites this paper
• An efficient cross-view image fusion method based on selected state space and hashing for promoting urban perception

  2024cites this paper
• SpaGBOL: Spatial-Graph-Based Orientated Localisation

  2024cites this paper
• Weakly-supervised Camera Localization by Ground-to-satellite Image Registration

  2024cites this paper
• UAV Geo-Localization for Navigation: A Survey

  2024cites this paper
• BEV-CV: Birds-Eye-View Transform for Cross-View Geo-Localisation

  2023cites this paper
• Simple, Effective and General: A New Backbone for Cross-view Image Geo-localization

  2023cites this paper
• Boosting 3-DoF Ground-to-Satellite Camera Localization Accuracy via Geometry-Guided Cross-View Transformer

  2023cites this paper
• Real‐time mosaic of multiple fisheye surveillance videos based on geo‐registration and rectification

  2023cites this paper
• Image-based Geolocalization by Ground-to-2.5D Map Matching

  2023cites this paper
• Cross-View Object Geo-Localization in a Local Region With Satellite Imagery

  2023cites this paper
• Geographic Semantic Network for Cross-View Image Geo-Localization

  2022influential citation
• Road Extraction Assisted Offset Regression Method in Cross-view Image- based Geo-localization

  2022cites this paper
• It’s Okay to Be Wrong: Cross-View Geo-Localization With Step-Adaptive Iterative Refinement

  2022cites this paper
• Beyond Geo-localization: Fine-grained Orientation of Street-view Images by Cross-view Matching with Satellite Imagery

  2022cites this paper
• SMDT: Cross-View Geo-Localization with Image Alignment and Transformer

  2022cites this paper
• Transformer-Guided Convolutional Neural Network for Cross-View Geolocalization

  2022cites this paper
• Beyond Cross-view Image Retrieval: Highly Accurate Vehicle Localization Using Satellite Image

  2022cites this paper
• TransGeo: Transformer Is All You Need for Cross-view Image Geo-localization

  2022cites this paper
• Vision-Based UAV Self-Positioning in Low-Altitude Urban Environments

  2022cites this paper
• Cross-view Geo-localization with Layer-to-Layer Transformer

  2021cites this paper
• Cross-view Geo-localization with Evolving Transformer

  2021cites this paper
• Where Am I Looking At? Joint Location and Orientation Estimation by Cross-View Matching

  2020cites this paper
• VIGOR: Cross-View Image Geo-localization beyond One-to-one Retrieval

  2020cites this paper
• Deep Neural Networks for Ortophoto-Based Vehicle Localization

  2020cites this paper
• Ground-to-Aerial Viewpoint Localization via Landmark Graphs Matching

  2020cites this paper
• Learning Discriminative Features for Image Retrieval

  2019cites this paper