Fast modified Horn & Schunck method for the estimation of optical flow fields

Abstract

This paper presents a fast , accurate and reliable modified Horn & Schunck approach for robust boundary preserving estimation of optical flow where only 30 iterations are used . The proposed method is derived from the benchmark algorithm of Horn & Schunck and Simoncelli's matched-pair 5 tap filters, such that it produces robust, fast and exact detection of motion boundaries and it is very simple to implement. Experimental results using synthetic and real optical flows are presented to demonstrate the effectiveness of our method in comparison to selected methods .

Research on UAV video motion target detection method based on optical flow network

“…Traditional methods for UAV video motion target detection include frame difference method, background modeling method, optical flow method, and deep learning method [1] . For example, Lucas et al proposed the Lucas-Kanade algorithm for computing dense optical flow [2] ; Barrir et al proposed the optical flow method by adding a momentum term to speed up the convergence of the algorithm [3] ; Liu Hongbin et al similarly proposed a weighted adaptive optical flow algorithm to shorten the running time of the optical flow method, which detects the motion target significantly and satisfies the optimal and suboptimal performance of the optical flow method [4] . The optical flow method has a better detection effect, but the achieved effect is limited to some specific scenes, and the detected motion targets are prone to missing content [5] .…”