HOME MISSION AND RESEARCH PUBLICATIONS HISTORY PEOPLE SEMINARS COURSES VIDEO ARCHIVE CONTACT

Jack Culpepper
EECS/Redwood Center, UC Berkeley

Dissertation talk: Learned factorization models to explain variability in natural image sequences

Thursday 12th of May 2011 at 12:00pm
508-20 Evans

Robust object recognition requires computational mechanisms that compensate for variability in the appearance of objects under natural viewing conditions. Yet, these have proven to be difficult to engineer. For this reason, the development of computational models that achieve invariance to the types of transformations that occur during natural viewing will both benefit our understanding of biological systems and help to achieve the goals of computer vision. This talk presents models that learn low dimensional representations of the transformations occurring in dynamic natural scenes. Good models of these transformations allow their effect to be compensated for through an inference process, which jointly estimates a stable percept and a parsimonious description of its appearance. My models are based on the idea of factoring apart image sequences into two types of latent variables: one that is relatively constant in time (the presence of a particular object), and another that gives a low dimensional time-varying representation of its appearance. Such a two component model is a general mechanism for teasing apart the causes that conspire to produce a time-varying image. When both components are represented by linear expansions, the resulting bilinear model can achieve some degree of image stabilization by utilizing the transformation model to explain the translation motions that occur in a small window of a movie. Yet, the recovered latent factors exhibit dependencies that motivate the investigation of a richer, exponential map as a second model for the dynamics of appearance. In addition to the translation motions captured by the linear appearance model, this richer model learns transformations that can compensate for rotations, expansions, and complex distortions in the data.


Join Email List

You can subscribe to our weekly seminar email list by sending an email to majordomo@lists.berkeley.edu that contains the words subscribe redwood in the body of the message.
(Note: The subject line can be arbitrary and will be ignored)