Multi-view face tracking video sequences

We provide datasets used in evaluating tracking using Multiplicative Kernels. The kernel formulation is described here:

Quan Yuan, Ashwin Thangali, Vitaly Ablavsky and Stan Sclaroff ,
"Multiplicative Kernels: Object Detection, Segmentation and Pose Estimation," CVPR, 2008.

The tracking formulation is described in a submission to IEEE T-PAMI.

(xvid video format)

             

     The face exemplar on top of each detection window indicates face orientation estimate. (sequence 1, sequence2)