Presenters:
Tom Polk, Stephen MacNeil and Jialei Li
TenniVis: Visualization for Tennis Match Analysis
In this talk, Tom will introduce TenniVis, a new tennis match visualization system designed for non-professional tennis coaches and players. It is based entirely on data that can be easily collected, such as score, point outcomes, point lengths, service information, and match videos that can be captured by one consumer-level camera. Tom will also talk about the pilot studies he conducted with two tennis coaches.
Visualizing Short Term and Longitudinal Student Performance
Stephen will present initial results from a project to study the progression of knowledge acquisition for undergraduate students and to identify places for improvement based on student performance evaluation.
Probabilistic Data Fusion for Ridge Based Adaptive Surface Reconstruction
Jialei will present an improved framework for adaptive resolution terrain surface generation from 3D points sets collected by multiple complementary sensors. This research provides a very general way to generate terrain models from sensors. Measurements with different sensor models are merged in a probabilistic way using an octree data structure. A surface probability function is constructed which represents the probability that the space spanned by the corresponding voxel contains the surface. Ridges of the surface probability function are extracted and connected to build the surface mesh with triangles adaptive to the local discretization of space given by the octree. Since the integrated surface generated by the marching triangles algorithm may contain cracks and holes, several modifications are proposed to improve the quality of the resulting model. Examples from the integration of two LiDAR scans are presented to demonstrate the performance of our work.
