StairNet: A Computer Vision Dataset for Stair Recognition

Citation Author(s):
Andrew Garrett
Kurbis
University of Toronto
Brokoslaw
Laschowski
University of Toronto
Alex
Mihailidis
University of Toronto
Submitted by:
Brokoslaw Laschowski
Last updated:
Tue, 04/12/2022 - 14:12
DOI:
10.21227/12jm-e336
Links:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

Computer vision can be used by robotic leg prostheses and exoskeletons to improve high-level transitions between different locomotion modes (e.g., level-ground walking to stair ascent) through the prediction of future environmental states. Here we developed the StairNet dataset to support research and development in vision-based automated stair recognition. The dataset builds on the ExoNet database – the largest and most diverse open-source dataset of wearable camera images of real-world walking environments. StairNet contains ~515,000 labelled images from six of the twelve original ExoNet classes. These images were reclassified into four classes using new definitions from a computer vision perspective with the goal of increasing the accuracy of the cutoff points between classes. The dataset was manually parsed several times during annotation to reduce misclassification errors and remove images with large obstructions. The StairNet dataset opens new opportunities for environment-adaptive control of robotic leg prostheses and exoskeletons. 

Reference:

1. Kurbis AG, Laschowski B, and Mihailidis A. (2022). Stair Recognition for Robotic Exoskeleton Control using Computer Vision and Deep Learning. IEEE International Conference on Rehabilitation Robotics (ICORR). Accepted.

2. Kurbis AG, Laschowski B, and Mihailidis A. (2022). An Automated Stair Recognition System for Robotic Exoskeleton Control using Deep Learning. IEEE International Conference on Robotics and Automation (ICRA). Accepted. 

Instructions: 

*Details regarding the StairNet dataset are provided in the ReadMe file. TFRecords dataset format is available upon request. Please email A. Garrett Kurbis (garrett.kurbis@utoronto.ca) or Dr. Brokoslaw Laschowski (brock.laschowski@mail.utoronto.ca) for any additional questions and/or technical assistance.