We developed a robust multiscale visual tracker of multiple objects in video using the Dual-tree Complex Wavelet Transform (DT-CWT). Real-valued wavelet transforms were previously used for visual tracking, but most suffer from shift variance and lack of directional selectivity. Therefore, we used DT-CWT to avoid such shortcomings. In our tracker, a captured video frame was represented as different subbands using DT-CWT. Then we applied N independent particle filters to a small subset of these subbands, where the choice of these subbands changed adaptively with each captured frame.
Finally, we fused the position tracks resulting from these particle filters to obtain final position tracks of multiple moving objects in the video. To demonstrate robustness of our visual tracker, we compared the performance of our multiscale tracker to a standard particle filter full resolution-based tracker and a single wavelet subband (LL)2 based tracker, our multiscale tracker demonstrates significantly better tracking performance.