While the concept of optical flow trajectories have

In our method, each image frame of the original video shot is replaced by a simplified image, called an MVFI (Motion Vector Flow Instances) motion template, that MK-2461 extracts the direction and strength of the velocity flow field (Díaz-Pereira et al., 2014 and Olivieri et al., 2012), found by frame-to-frame differencing. Each template image can be further projected as a point into a reduced dimensional eigenspace through a Principal component analysis (PCA), or kernel-PCA (KPCA) transformation, so that frames of the entire video sequence, when projected into autosomes space, trace out a unique curve we call the covariant spatio-temporal trajectory.