Event Location: NSH 1305

Abstract: *** Talk 1 ***

Spatiotemporal data is inherent in a number of graphics applications including animation, simulation, and object and camera tracking. The principal modes of variation in the spatial geometry of objects are typically modeled using dimensionality reduction techniques, while concurrently, trajectory representations like splines and autoregressive models are widely used to exploit the temporal regularity of deformation. In this talk, we present the bilinear spatiotemporal basis as a model that simultaneously exploits spatial and temporal regularity while maintaining the ability to generalize well to new sequences. This factorization allows the use of analytical, predefined functions to represent temporal variation (e.g., B-Splines or the Discrete Cosine Transform) resulting in efficient model representation and estimation. We apply the bilinear model to natural spatiotemporal phenomena, including face, body, and cloth motion data, and compare it in terms of compaction, generalization ability, predictive precision, and efficiency to existing models. We demonstrate the application of the model to a number of graphics tasks including labeling, gap-filling, de-noising, and motion touch-up.

Project webpage: http://graphics.cs.cmu.edu/projects/bilinearbasis/

*** Talk 2 ***

I’ll talk a bit about the 3D sensing facility we’re building in the basement of NSH. The studio is a geodesic dome with nearly 500 synchronized cameras, along with projectors and eventually time-of-flight cameras. The goal of the facility is to push the accuracy of 3D reconstruction of dynamic behavior to the next plateau of performance. At the conclusion of the short presentation, I’ll provide a short tour of the studio.