Title:  RAPTr: Robust Articulated Point-set Tracking

Committee:

Dr. Patricio Vela, ECE, Chair , Advisor

Dr. Ayanna Howard, ECE

Dr. Anthony Yezzi, ECE

Dr. Matthieu Bloch, ECE

Dr. Yu-Ping Chen, GSU

Abstract:

The objective of this work is to present the Robust Articulated Point-set Tracking (RAPTr) system. It works by synthesizing components from articulated model-based and machine learning methods in a framework for pose estimation. Purely machine learning based pose estimation methods are robust to image artifacts. However, they require large annotated datasets. On the other hand, articulated model-based methods can emulate an infinite number of poses while respecting the subject's geometry but are susceptible to local minimum, as they are sensitive to the various artifacts that appear in realistic imaging conditions (e.g. subtle background noise due to shadows or movements). The proposed work outlines a method that combines aspects from both machine learning and articulated model-based fitting in a manner that exploits the benefits both approaches provide. When necessary, an intermediate representation is defined so that the two approaches may operate on compatible inputs. The proposed solution will be applied to articulated pose estimation problems where the pose estimate accuracy is the priority.