# DROW: Real-Time Deep Learning based Wheelchair Detection in 2D Range Data

arXiv: https://arxiv.org/abs/1603.02636 · HTML: https://mlsanity.com/paper/1603.02636/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Mar 2016.

## Built on by (most cited first, 3 of 3)

- [2D vs. 3D LiDAR-based Person Detection on Mobile Robots](https://mlsanity.com/paper/2106.11239/) (2021)
- [Disability Representations: Finding Biases in Automatic Image Generation](https://mlsanity.com/paper/2406.14993/) (2024)
- [MARF: Multiscale Adaptive-switch Random Forest for Leg Detection with 2D Laser Scanners](https://mlsanity.com/paper/2204.06833/) (2022)

## Abstract (arXiv)

We introduce the DROW detector, a deep learning based detector for 2D range data. Laser scanners are lighting invariant, provide accurate range data, and typically cover a large field of view, making them interesting sensors for robotics applications. So far, research on detection in laser range data has been dominated by hand-crafted features and boosted classifiers, potentially losing performance due to suboptimal design choices. We propose a Convolutional Neural Network (CNN) based detector for this task. We show how to effectively apply CNNs for detection in 2D range data, and propose a depth preprocessing step and voting scheme that significantly improve CNN performance. We demonstrate our approach on wheelchairs and walkers, obtaining state of the art detection results. Apart from the training data, none of our design choices limits the detector to these two classes, though. We provide a ROS node for our detector and release our dataset containing 464k laser scans, out of which 24k were annotated.