# VIRD: View-Invariant Representation through Dual-Axis Transformation for Cross-View Pose Estimation

arXiv: https://arxiv.org/abs/2603.12918 · HTML: https://mlsanity.com/paper/2603.12918/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 36 references that are in the map.
- First posted on arXiv 13 Mar 2026.

## Builds on (most cited first, 23 of 23)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [VIGOR: Cross-View Image Geo-localization beyond One-to-one Retrieval](https://mlsanity.com/paper/2011.12172/) (2020)
- [Where am I looking at? Joint Location and Orientation Estimation by Cross-View Matching](https://mlsanity.com/paper/2005.03860/) (2020)
- [Beyond Cross-view Image Retrieval: Highly Accurate Vehicle Localization Using Satellite Image](https://mlsanity.com/paper/2204.04752/) (2022)
- [OrienterNet: Visual Localization in 2D Public Maps with Neural Matching](https://mlsanity.com/paper/2304.02009/) (2023)
- [Fine-Grained Cross-View Geo-Localization Using a Correlation-Aware Homography Estimator](https://mlsanity.com/paper/2308.16906/) (2023)
- [Visual Cross-View Metric Localization with Dense Uncertainty Estimates](https://mlsanity.com/paper/2208.08519/) (2022)
- [SliceMatch: Geometry-guided Aggregation for Cross-View Pose Estimation](https://mlsanity.com/paper/2211.14651/) (2022)
- [Boosting 3-DoF Ground-to-Satellite Camera Localization Accuracy via Geometry-Guided Cross-View Transformer](https://mlsanity.com/paper/2307.08015/) (2023)
- [A Survey on Global LiDAR Localization: Challenges, Advances and Open Problems](https://mlsanity.com/paper/2302.07433/) (2023)
- [Convolutional Cross-View Pose Estimation](https://mlsanity.com/paper/2303.05915/) (2023)
- [FG2: Fine-Grained Cross-View Localization by Fine-Grained Feature Matching](https://mlsanity.com/paper/2503.18725/) (2025)
- [Learning Dense Flow Field for Highly-accurate Cross-view Camera Localization](https://mlsanity.com/paper/2309.15556/) (2023)
- [Satellite Image Based Cross-view Localization for Autonomous Vehicle](https://mlsanity.com/paper/2207.13506/) (2022)
- [(LC)2: LiDAR-Camera Loop Constraints For Cross-Modal Place Recognition](https://mlsanity.com/paper/2304.08660/) (2023)
- [View Consistent Purification for Accurate Cross-View Localization](https://mlsanity.com/paper/2308.08110/) (2023)
- [PIDLoc: Cross-View Pose Optimization Network Inspired by PID Controllers](https://mlsanity.com/paper/2503.02388/) (2025)
- [Relatively Lazy: Indoor-Outdoor Navigation Using Vision and GNSS](https://mlsanity.com/paper/2101.05107/) (2021)

## Abstract (arXiv)

Accurate global localization is critical for autonomous driving and robotics, but GNSS-based approaches often degrade due to occlusion and multipath effects. As an emerging alternative, cross-view pose estimation predicts the 3-DoF camera pose corresponding to a ground-view image with respect to a geo-referenced satellite image. However, existing methods struggle to bridge the significant viewpoint gap between the ground and satellite views mainly due to limited spatial correspondences. We propose a novel cross-view pose estimation method that constructs view-invariant representations through dual-axis transformation (VIRD). VIRD first applies a polar transformation to the satellite view to facilitate horizontal correspondence, then uses context-enhanced positional attention on the ground and polar-transformed satellite features to mitigate vertical misalignment, explicitly bridging the viewpoint gap. To further strengthen view invariance, we introduce a view-reconstruction loss that encourages the derived representations to reconstruct the original and cross-view images. Experiments on the KITTI and VIGOR datasets demonstrate that VIRD outperforms the state-of-the-art methods without orientation priors, reducing median position and orientation errors by 50.7% and 76.5% on KITTI, and 18.0% and 46.8% on VIGOR, respectively.