Visual Tracking by Affine Kernel Fitting Using Color and Object Boundary

Kernel-based trackers aggregate image features within the support of a kernel (a mask) regardless of their spatial structure. These trackers spatially fit the kernel (usually in location and in scale) such that a function of the aggregate is optimized. We propose a kernel-based visual tracker that exploits the constancy of color and the presence of color edges along the target boundary. The tracker estimates the best affinity of a spatially aligned pair of kernels, one of which is color-related and the other of which is object boundary-related. In a sense, this work extends previous kernel-based track-ers by incorporating the object boundary cue into the track-ing process and by allowing the kernels to be affinely trans-formed instead of only translated and isotropically scaled. These two extensions make for more precise target local-ization. Moreover, a more accurately localized target facil-itates safer updating of its reference color model, further enhancing the tracker’s robustness. The improved tracking is demonstrated for several challenging image sequences. 1