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. (a) (b)

In this paper we describe a citizen science system for solving time-consuming and labor-intensive problems, using crowdsourcing and efficient geometric algorithms. Specifically, the system can be used to trace static objects in images (such as trees in an urban environment), or to generate trajectories of moving objects in videos (such as ants in an ant colony). The traces of the static objects can provide quantitative measurements such as size, shape and appearance, for example in monitoring the health of the trees in New York City’s Million Trees Initiative. It is relatively easy to plant a million trees, but ensuring they are healthy and taken care of is a challenge on a different scale, and a challenge where citizen scientists can make a big difference. The ant trajectories extracted from videos of ant colonies are needed by biologists studying longitudinal behavioral patterns in insect colonies. Existing automated solutions are not good enough, and there is only so much data that even motivated students can annotate in the research lab. AngryAnts is our on-line application which displays short video segments, specifies which ant needs to be traced and allows the citizen scientist to enter the trajectory in a first-person shooter style via mouse clicks. Submitted trajectories are verified using a ReCaptcha method, where part of the trajectory

Road networks are important datasets for an increasing number of applications. However, the creation and maintenance of such datasets pose interesting research challenges. This work proposes an automatic road network generation algorithm that takes vehicle tracking data in the form of trajectories as input and produces a road network graph. This effort addresses the challenges of evolving map data sets, specifically by focusing on (i) automatic map-attribute generation (weights), (ii) automatic road network generation, and (iii) by providing a quality assessment. An experimental study assesses the quality of the algorithms by generating a part of the road network of Athens, Greece, using trajectories derived from GPS tracking a school bus fleet.