# EDENet: Echo Direction Encoding Network for Place Recognition Based on Ground Penetrating Radar

arXiv: https://arxiv.org/abs/2502.20643 · HTML: https://mlsanity.com/paper/2502.20643/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 33 references that are in the map.
- First posted on arXiv 28 Feb 2025.

## Builds on (most cited first, 8 of 8)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [OverlapNet: Loop Closing for LiDAR-based SLAM](https://mlsanity.com/paper/2105.11344/) (2021)
- [SeqOT: A Spatial-Temporal Transformer Network for Place Recognition Using Sequential LiDAR Data](https://mlsanity.com/paper/2209.07951/) (2022)
- [Learning to Remove Clutter in Real-World GPR Images Using Hybrid Data](https://mlsanity.com/paper/2205.08135/) (2022)
- [gprHOG and the popularity of Histogram of Oriented Gradients (HOG) for Buried Threat Detection in Ground-Penetrating Radar](https://mlsanity.com/paper/1806.01349/) (2018)

## Abstract (arXiv)

Ground penetrating radar (GPR) based localization has gained significant recognition in robotics due to its ability to detect stable subsurface features, offering advantages in environments where traditional sensors like cameras and LiDAR may struggle. However, existing methods are primarily focused on small-scale place recognition (PR), leaving the challenges of PR in large-scale maps unaddressed. These challenges include the inherent sparsity of underground features and the variability in underground dielectric constants, which complicate robust localization. In this work, we investigate the geometric relationship between GPR echo sequences and underground scenes, leveraging the robustness of directional features to inform our network design. We introduce learnable Gabor filters for the precise extraction of directional responses, coupled with a direction-aware attention mechanism for effective geometric encoding. To further enhance performance, we incorporate a shift-invariant unit and a multi-scale aggregation strategy to better accommodate variations in di-electric constants. Experiments conducted on public datasets demonstrate that our proposed EDENet not only surpasses existing solutions in terms of PR performance but also offers advantages in model size and computational efficiency.