1. Taylor Rassmann

2.  Action is often distinguished from activity in the sense that action is an individual atomic unit of activity. In particular, human action refers to physical body motion.

3.  Novel feature representation that is derived from space-time local (steering) regression kernels (3D LSKs)  This feature representation can attain data even in instances of high distortion and data uncertainty

4.  This is achieved by measuring the likeness of a voxel to its surroundings based on computation of a distance between points.  These points are measured (along the shortest path) on a manifold defined by the embedding of the video data in 4D  For better classification performance, space time saliency detection is applied to larger videos to crop to a shorter action clip

5.  The key idea behind 3D LSKs is to robustly obtain local space-time geometric structures by analyzing the photometric (voxel value) differences based on estimated space-time gradients, and use this structure information to determine the shape and size of a canonical kernel (descriptor).

6.  Use of pair wise distances of salient regions  Saliency extraction complete from a few actions in KTH data set

7.  Code for saliency statistics in progress  This will implement the distance metrics for pair wise distances between features  Possible use of thresholding for some of the salient regions to prevent merging  This will help acquire different parts of features  Note: Thresholding must be careful not to be to high or low to eliminate some of the important data