# JRN-Geo: A Joint Perception Network based on RGB and Normal images for Cross-view Geo-localization

arXiv: https://arxiv.org/abs/2509.05696 · HTML: https://mlsanity.com/paper/2509.05696/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 34 references that are in the map.
- First posted on arXiv 6 Sep 2025.

## Builds on (most cited first, 19 of 19)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction](https://mlsanity.com/paper/2206.00665/) (2022)
- [Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scans](https://mlsanity.com/paper/2110.04994/) (2021)
- [DeepMVS: Learning Multi-view Stereopsis](https://mlsanity.com/paper/1804.00650/) (2018)
- [Lending Orientation to Neural Networks for Cross-view Geo-localization](https://mlsanity.com/paper/1903.12351/) (2019)
- [University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization](https://mlsanity.com/paper/2002.12186/) (2020)
- [Sample4Geo: Hard Negative Sampling For Cross-View Geo-Localisation](https://mlsanity.com/paper/2303.11851/) (2023)
- [Learning Feature Descriptors using Camera Pose Supervision](https://mlsanity.com/paper/2004.13324/) (2020)
- [Each Part Matters: Local Patterns Facilitate Cross-view Geo-localization](https://mlsanity.com/paper/2008.11646/) (2020)
- [SUES-200: A Multi-height Multi-scene Cross-view Image Benchmark Across Drone and Satellite](https://mlsanity.com/paper/2204.10704/) (2022)
- [A Transformer-Based Feature Segmentation and Region Alignment Method For UAV-View Geo-Localization](https://mlsanity.com/paper/2201.09206/) (2022)
- [VEGS: View Extrapolation of Urban Scenes in 3D Gaussian Splatting using Learned Priors](https://mlsanity.com/paper/2407.02945/) (2024)
- [Learning Cross-view Geo-localization Embeddings via Dynamic Weighted Decorrelation Regularization](https://mlsanity.com/paper/2211.05296/) (2022)
- [Surface Normals in the Wild](https://mlsanity.com/paper/1704.02956/) (2017)

## Built on by (most cited first, 2 of 2)

- [(MGS)2-Net: Unifying Micro-Geometric Scale and Macro-Geometric Structure for Cross-View Geo-Localization](https://mlsanity.com/paper/2602.10704/) (2026)
- [InfoGeo: Information-Theoretic Object-Centric Learning for Cross-View Generalizable UAV Geo-Localization](https://mlsanity.com/paper/2605.07099/) (2026)

## Abstract (arXiv)

Cross-view geo-localization plays a critical role in Unmanned Aerial Vehicle (UAV) localization and navigation. However, significant challenges arise from the drastic viewpoint differences and appearance variations between images. Existing methods predominantly rely on semantic features from RGB images, often neglecting the importance of spatial structural information in capturing viewpoint-invariant features. To address this issue, we incorporate geometric structural information from normal images and introduce a Joint perception network to integrate RGB and Normal images (JRN-Geo). Our approach utilizes a dual-branch feature extraction framework, leveraging a Difference-Aware Fusion Module (DAFM) and Joint-Constrained Interaction Aggregation (JCIA) strategy to enable deep fusion and joint-constrained semantic and structural information representation. Furthermore, we propose a 3D geographic augmentation technique to generate potential viewpoint variation samples, enhancing the network's ability to learn viewpoint-invariant features. Extensive experiments on the University-1652 and SUES-200 datasets validate the robustness of our method against complex viewpoint ariations, achieving state-of-the-art performance.