Home > Published Issues > 2021 > Volume 9, No. 4, December 2021 >

People Detection with Depth Silhouettes and Convolutional Neural Networks on a Mobile Robot

Florian Spiess 1, Lucas Reinhart 2, Norbert Strobel 2, Dennis Kaiser 3, Samuel Kounev 3, and Tobias Kaupp 4
1. Faculty of Electrical Engineering, University of Applied Sciences Wuerzburg-Schweinfurt, Schweinfurt, Germany
2. Institute of Medical Engineering, University of Applied Sciences Wuerzburg-Schweinfurt, Schweinfurt, Germany
3. Faculty of Mathematics and Computer Science, Julius-Maximilians-University of Wuerzburg, Wuerzburg, Germany
4. University of Applied Sciences Wuerzburg-Schweinfurt, Schweinfurt, Germany

Abstract—This paper presents a novel people detection approach for mobile robot applications based on a combination of classical computer vision techniques and a state-of-the-art neural network. Our approach involves an RGB-D camera as an environmental sensor. The depth data is used to extract silhouettes around people. The RGB images are subsequently augmented with this border information before passing it to the neural network. Under challenging lighting conditions, our system was able to outperform the neural network trained on regular RGB data alone by a factor of two.

Index Terms—neural network, mobile robot

Cite: Florian Spiess, Lucas Reinhart, Norbert Strobel, Dennis Kaiser, Samuel Kounev, and Tobias Kaupp, "People Detection with Depth Silhouettes and Convolutional Neural Networks on a Mobile Robot," Journal of Image and Graphics, Vol. 9, No. 4, pp. 135-139, December 2021. doi: 10.18178/joig.9.4.135-139