Abstract

The Lucas-Kanade (LK) method is a classic tracking algorithm ex- ploiting target structural constraints thorough template matching. Extended Lucas Kanade or ELK casts the original LK algorithm as a maximum likelihood opti- mization and then extends it by considering pixel object / background likelihoods in the optimization. Template matching and pixel-based object / background seg- regation are tied together by a uni fied Bayesian framework. In this framework two log-likelihood terms related to pixel object / background affiliation are introduced in addition to the standard LK template matching term. Tracking is performed us- ing an EM algorithm, in which the E-step corresponds to pixel object/background inference, and the M-step to parameter optimization. The final algorithm, imple- mented using a classi fier for object / background modeling and equipped with simple template update and occlusion handling logic, is evaluated on two chal- lenging data-sets containing 50 sequences each. The first is a recently published benchmark where ELK ranks 3rd among 30 tracking methods evaluated. On the second data-set of vehicles undergoing severe view point changes ELK ranks in 1st place outperforming state-of-the-art methods.