# On the descriptive power of LiDAR intensity images for segment-based loop closing in 3-D SLAM

arXiv: https://arxiv.org/abs/2108.01383 · HTML: https://mlsanity.com/paper/2108.01383/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 24 references that are in the map.
- First posted on arXiv 3 Aug 2021.

## Builds on (most cited first, 11 of 11)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [SuperPoint: Self-Supervised Interest Point Detection and Description](https://mlsanity.com/paper/1712.07629/) (2017)
- [Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks](https://mlsanity.com/paper/1910.01279/) (2019)
- [PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition](https://mlsanity.com/paper/1804.03492/) (2018)
- [OverlapNet: Loop Closing for LiDAR-based SLAM](https://mlsanity.com/paper/2105.11344/) (2021)
- [PCAN: 3D Attention Map Learning Using Contextual Information for Point Cloud Based Retrieval](https://mlsanity.com/paper/1904.09793/) (2019)
- [SegMatch: Segment based loop-closure for 3D point clouds](https://mlsanity.com/paper/1609.07720/) (2016)
- [SegMap: Segment-based mapping and localization using data-driven descriptors](https://mlsanity.com/paper/1909.12837/) (2019)
- [Local Descriptor for Robust Place Recognition using LiDAR Intensity](https://mlsanity.com/paper/1811.12646/) (2018)
- [Learning to See the Wood for the Trees: Deep Laser Localization in Urban and Natural Environments on a CPU](https://mlsanity.com/paper/1902.10194/) (2019)
- [Spatiotemporal Calibration of Camera and 3D Laser Scanner](https://mlsanity.com/paper/2006.16081/) (2020)

## Built on by (most cited first, 1 of 1)

- [MinkLoc3D-SI: 3D LiDAR place recognition with sparse convolutions, spherical coordinates, and intensity](https://mlsanity.com/paper/2112.06539/) (2021)

## Abstract (arXiv)

We propose an extension to the segment-based global localization method for LiDAR SLAM using descriptors learned considering the visual context of the segments. A new architecture of the deep neural network is presented that learns the visual context acquired from synthetic LiDAR intensity images. This approach allows a single multi-beam LiDAR to produce rich and highly descriptive location signatures. The method is tested on two public datasets, demonstrating an improved descriptiveness of the new descriptors, and more reliable loop closure detection in SLAM. Attention analysis of the network is used to show the importance of focusing on the broader context rather than only on the 3-D segment.