Asymmetric Bilateral Phase Correlation for Optical Flow Estimation in the Frequency Domain
This addresses optical flow estimation for computer vision applications, but appears incremental as it builds on existing phase correlation methods.
The paper tackles motion estimation in images by extending phase correlation to handle multiple motions in the frequency domain, using a novel Bilateral-Phase Correlation technique that preserves motion boundaries, and reports outperforming recent state-of-the-art phase correlation based optical flow methods on well-known datasets.
We address the problem of motion estimation in images operating in the frequency domain. A method is presented which extends phase correlation to handle multiple motions present in an area. Our scheme is based on a novel Bilateral-Phase Correlation (BLPC) technique that incorporates the concept and principles of Bilateral Filters retaining the motion boundaries by taking into account the difference both in value and distance in a manner very similar to Gaussian convolution. The optical flow is obtained by applying the proposed method at certain locations selected based on the present motion differences and then performing non-uniform interpolation in a multi-scale iterative framework. Experiments with several well-known datasets with and without ground-truth show that our scheme outperforms recently proposed state-of-the-art phase correlation based optical flow methods.