We propose an innovative approach for human activity recognition based on affine-invariant shape representation and SVMbased\r\nfeature classification. In this approach, a compact computationally efficient affine-invariant representation of action shapes\r\nis developed by using affine moment invariants. Dynamic affine invariants are derived from the 3D spatiotemporal action volume\r\nand the average image created fromthe 3D volume and classified by an SVM classifier. On two standard benchmark action datasets\r\n(KTH andWeizmann datasets), the approach yields promising results that compare favorably with those previously reported in the\r\nliterature, while maintaining real-time performance.
Loading....