On line predictive appearance-based tracking

Abstract

We present a novel predictive statistical framework to improve the performance of an EigenTracker. In addition, we use fast and efficient eigenspace updates to learn new views of the object being tracked on the fly. We also incorporate a new Importance Sampling mechanism which increases the robustness of the EigenTracker, and enables it to track nonconvex objects better. Our EigenTracker is flexible -it is possible to use it symbiotically with other trackers. We show its successful application in hand gesture analysis; and face and person tracking.

Online Improved Eigen Tracking

“…The method can robustly track an object in the presence of large viewpoint changes, partial occlusion, lighting variation, changes to the shape of the object shaky cameras, and motion blur. Moreover avoidance of non-linear optimization makes our tracking task faster than that of [7]. …”

“…The state estimate is used to generate the predictions for the next frame. The prediction framework we used is motivated by predictive Eigen tracker [7].…”

“…The Isard and Blake CONDENSATION algorithm [2] can represent simultaneous multiple hypothesis. There can be several ways by virtue of which the power of EigenTracker and particle filter can be combined like [7] and [8]. But these have the overhead of non-linear optimization.…”

Hand gesture modelling and recognition involving changing shapes and trajectories, using a Predictive EigenTracker

“…This is an important consideration for an appearance-based method, since we do not want much background to be learnt as part of the eigenspace representation of the object. Our Predictive EigenTracker (Gupta et al, 2004) augments the capability of an EigenTracker in three ways. One of the main factors for the inefficiency of the EigenTracker is the absence of a predictive framework.…”

“…Further, our method is independent of common hand shape deformations: rotation, translation, scale and shear. (Our system works on top of our Predictive EigenTracker (Gupta et al, 2004), an enhancement of the original EigenTracker (Black and Jepson, 1998). This tracker gives us both appearance as well as motion/trajectory information, and is robust to background clutter and structured noise.…”