We report on our experience building and implementing a robotics course for Computer Science based on the LEGO NXT. The goal of the course is to provide students with an opportunity to learn the fundamental principles involved in designing, programming, and operating mobile robots in a completely experiential setting, while abstracting away the hardware complexity of typical robotics platforms. To this end, we built the course around the LEGO NXT robot kits. We have developed a software framework that allows the NXT to be used in conjunction with a laptop computer and a webcam to enable our students to learn in the context of a robotics soccer game. We describe the process of designing the robotics soccer project so that it is tightly coupled with course learning goals, and report on our experience running the course over the past three years. The software platform is freely available, and allows anyone with a webcam-equipped laptop and an NXT set to implement their own soccer playing bot.

We report on our experience implementing a Teaching Assistant Training Program for Computer Science Teaching Assistants. The program is rooted in the well-established Instructional Skills Workshop's principles and methodologies. We describe the structure of the program, reflect on its successes, and discuss future expansion of, and improvements to the program.

With growing enrolment and ongoing research into best practices for team work, many instructors are rethinking how to form, evaluate, and manage teams. In this BoF, instructors will discuss the strategies they have employed, the effectiveness of those approaches, and the tools that support administering teams. Discussion topics may include self- and peer-evaluation, grading strategies, software support, and conflict resolution.

In this paper we focus on the problem of ranking news stories within their historical context by exploiting their content similarity. We observe that news stories evolve and thus have to be ranked in a time and query dependent manner. We do this in two steps. First, the mining step discovers metastories, which constitute meaningful groups of similar stories that occur at arbitrary points in time. Second, the ranking step uses well known measures of content similarity to construct implicit links among all metastories, and uses them to rank those metastories that overlap the time interval provided in a user query. We use real data from conventional and social media sources (weblogs) to study the impact of different meta-aggregation techniques and similarity measures in the final ranking. We evaluate the framework using both objective and subjective criteria, and discuss the selection of clustering method and similarity measure that lead to the best ranking results.

We present a novel algorithm for image denoising. Our algorithm is based on random walks over arbitrary neighbourhoods surrounding a given pixel. The size and shape of each neighbourhood are determined by the configuration and similarity of nearby pixels. Assuming that pixels within the neighbourhood of x0 are likely to have been generated by the same random process, we want the weights used to mix these pixels during denoising to depend on the similarity between them and x0. At the same time, we require the random walk to follow a smooth path from x0 to any other pixel in the neighbourhood, so the transition probabilities should also depend on the similarity between pairs of neighbouring pixels along any given path. With this in mind, we define a random walk originating at pixel x0 as an ordered sequence of pixels T0,k = {x0,x1, . . . ,xk} visited along the path from x0 to xk. Within this sequence, the probability of a transition between two consecutive pixels x j and x j+1 is defined to be

Detection of visually salient image regions is useful for applications like object segmentation, adaptive compression, and object recognition. In this paper, we introduce a method for salient region detection that outputs full resolution saliency maps with well-defined boundaries of salient objects. These boundaries are preserved by retaining substantially more frequency content from the original image than other existing techniques. Our method exploits features of color and luminance, is simple to implement, and is computationally efficient. We compare our algorithm to five state-of-the-art salient region detection methods with a frequency domain analysis, ground truth, and a salient object segmentation application. Our method outperforms the five algorithms both on the ground-truth evaluation and on the segmentation task by achieving both higher precision and better recall.

We present an algorithm for clustering sets of detected interest points into groups that correspond to visually distinct structure. Through the use of a suitable colour and texture representation, our clustering method is able to identify keypoints that belong to separate objects or background regions. These clusters are then used to constrain the matching of keypoints over pairs of images, resulting in greatly improved matching under difficult conditions. We present a thorough evaluation of each component of the algorithm, and show its usefulness on difficult matching problems.

We present a quantitative evaluation of SE-MinCut, a novel segmentation algorithm based on spectral embedding and minimum cut. We use human segmentations from the Berkeley segmentation database as ground truth and propose suitable measures to evaluate segmentation quality. With these measures we generate precision/recall curves for SE-MinCut and three of the leading segmentation algorithms: mean-shift, normalized Cuts, and the local variation algorithm. These curves characterize the performance of each algorithm over a range of input parameters. We compare the precision/recall curves for the four algorithms and show segmented images that support the conclusions obtained from the quantitative evaluation.

Recently it has been shown that min-cut algorithms can provide perceptually salient image segments when they are given appropriate proposals for source and sink regions. Here we explore the use of random walks and associated spectral embedding techniques for the automatic generation of suitable proposal regions. To do this, we first derive a mathematical connection between spectral embedding and anisotropic image smoothing kernels. We then use properties of the spectral embedding and the associated smoothing kernels to select multiple pairs of source and sink regions for min-cut. This typically provides an over-segmentation, and therefore region merging is used to form the final image segmentation. We demonstrate this process on several sample images. 1