Citation Context

... already discusses a simple example. However,by introducing adequate objective functions,the optimization problem can acquire physical meaning in the context of a computer vision task. For example,in =-=[14]-=-,by defining the distance between the distributions of the model and a candidate of the target,nonrigid objects were tracked in an image sequence under severe distortions. The distance was defined at ...

Citation Context

...e can introduce a large bias in the estimated location of the target, and the resulting measure is scale variant (see [37, p. 262] for a discussion). We mention that since its original publication in =-=[18]-=-, the idea of kernel-based tracking has been exploited and developed forward by various researchers. Chen and Liu [14] experimented with the same kernel-weighted histograms, but employed the Kullback-...

Citation Context

...ods such as the mean-shift algorithm use viewpoint-insensitive object appearance models to track objects through non-rigid pose changes without any prior knowledge of scene structure or camera motion =-=[4]-=-. Kalman filter extensions achieve more robust tracking of maneuvering objects through the introduction of statistical models of object and camera motion [2]. Particle filtering extensions enable trac...

Citation Context

...exemplars [34], foreground/background models [16] templates with updating [26]; prediction schemes using particle filters [17], joint probabilistic data association filters [30], kernel-based filters =-=[8]-=- [11], support vector machines [2] [36] and variational inference [35]. These algorithms usually build or learn a model of the target object first and then use it for tracking, without adapting the mo...

Citation Context

...we define the candidate region in which we formulate the HSV histogram as R(xt) � lt + stW , then a kernel density estimate K(xt) � {k(n; xt)}n=1,...,N of the color distribution at time t is given=-= by [1, 14]: k(n; xt) =-=- η � d∈R(xt) δ[bt(d) − n] (1) where δ is the delta function, η is a normalizing constant which ensures k to be a probability distribution, �N n=1 k(n; xt) = 1, and a location d could be an...

Citation Context

...scaling, and object deformation. It is also relatively stable under rotation in depth in certain applications. Therefore, color distributions have been used successfully to track nonrigid bodies [5], =-=[26]��-=-�[28], e.g., for tracking heads [29], [28], [30], [27], hands [31], and other body parts against cluttered backgrounds from stationary or moving platforms. Color distributions have also been used for ...

Citation Context

...appearance model since this is usually the most challenging to design. Although many tracking methods employ static appearance models that are either defined manually or trained using the first frame =-=[16, 8, 1]-=-, these methods tend to have difficulties tracking objects that exhibit significant appearance changes. It has been shown that in many scenarios an adaptive appearance model, which evolves during the ...