|
Deep Learning for Human Part Discovery in Images
Gabriel L. Oliveira, Abhinav Valada, Claas Bollen, Wolfram Burgard and Thomas Brox
Proc. of the IEEE Int. Conf. on Robotics &Automation (ICRA)
2016
oliveira16icra.pdf
Notes:
This paper addresses the problem of human body part segmentation in conventional RGB images, which has
several applications in robotics, such as learning from demonstration and human-robot handovers. The proposed solution is based on Convolutional Neural Networks (CNNs). We present a network architecture that assigns each pixel to one of a predefined set of human body part classes, such as head, torso, arms, legs. After initializing weights with a very deep convolutional network for image classification, the network can be trained end-to-end and yields precise class predictions at the original input resolution. Our architecture particularly improves on over-fitting issues in the up-convolutional part of the network. Relying only on RGB rather than RGB-D images also allows us to apply the approach outdoors. The network achieves state-of-the-art performance on the PASCAL Parts dataset. Moreover, we introduce two new part segmentation datasets, the Freiburg sitting people dataset and the Freiburg people in disaster dataset. We also present results obtained with a ground robot and an unmanned aerial vehicle.
BibTeX:
@inproceedings{oliveira2016,
author = {Gabriel Oliveira and Abhinav Valada and Claas Bollen and Wolfram Burgard and Thomas Brox},
title = {Deep Learning for Human Part Discovery in Images},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
year = 2016,
url = {http://ais.informatik.uni-freiburg.de/publications/papers/oliveira16icra.pdf},
address = {Stockholm, Sweden}
}
|
|
Search for a specific document