A Statistical Model for Recreational Trails in Aerial Images

Andrew Predoehl, Scott Morris, Kobus Barnard; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2013, pp. 337-344

Abstract


We present a statistical model of aerial images of recreational trails, and a method to infer trail routes in such images. We learn a set of textons describing the images, and use them to divide the image into super-pixels represented by their texton. We then learn, for each texton, the frequency of generating on-trail and off-trail pixels, and the direction of trail through on-trail pixels. From these, we derive an image likelihood function. We combine that with a prior model of trail length and smoothness, yielding a posterior distribution for trails, given an image. We search for good values of this posterior using a novel stochastic variation of Dijkstra's algorithm. Our experiments, on trail images and groundtruth collected in the western continental USA, show substantial improvement over those of the previous best trail-finding method.

Related Material


[pdf]
[bibtex]
@InProceedings{Predoehl_2013_CVPR,
author = {Predoehl, Andrew and Morris, Scott and Barnard, Kobus},
title = {A Statistical Model for Recreational Trails in Aerial Images},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2013}
}