Abstract
Pedestrian pose prediction is an important topic related closely to robotics and automation. Accurate predictions of human pose and motion can facilitate a more thorough understanding and analysis of human behavior, which benefits real-world applications, such as human-robot interaction, humanoid and bipedal robot design, and safe navigation of mobile robots and autonomous vehicles. This article describes a deep predictive coding network-based approach for unsupervised pedestrian pose prediction from two-dimensional (2D) camera imagery and provides experimental results on two real-world autonomous vehicle data sets. This article also presents a discussion on topics for future work in unsupervised and semi-supervised pedestrian pose prediction and its potential applications in robotics and automation systems.
Links
Citation
Plain Text:
X. Du, R. Vasudevan and M. Johnson-Roberson, “Unsupervised Pedestrian Pose Prediction: A Deep Predictive Coding Network-Based Approach for Autonomous Vehicle Perception,” in IEEE Robotics & Automation Magazine, vol. 27, no. 2, pp. 129-138, June 2020,
doi: 10.1109/MRA.2020.2976313.
BibTex:
@ARTICLE{du2020unsupervised,
author={X. {Du} and R. {Vasudevan} and M. {Johnson-Roberson}},
journal={IEEE Robotics and Automation Magazine},
title={Unsupervised Pedestrian Pose Prediction: A Deep Predictive Coding Network-Based Approach for Autonomous Vehicle Perception},
year={2020},
volume={27},
number={2},
pages={129-138},
doi={10.1109/MRA.2020.2976313},
month={June}}