Improving Horn-Schunck

From LRDE

Revision as of 17:06, 9 January 2018 by Bot (talk | contribs) (Created page with "{{CSIReport | authors = Sylvain Lobry | title = Improving Horn-Schunck | year = 2012 | abstract = Computing optical flow has many applications including motion estimation, or ...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Abstract

Computing optical flow has many applications including motion estimation, or as a first step towards video inpainting. The optical flow equation can not be solved as such due to the aperture problem (two unknowns for one single equation). One set of algorithms tries to solve this equation based on a global strategy, which is, trying to have small variations in the flow. The first of them is Horn-Schunck. This algorithm, even though it globally works, has several drawbacks, including being slow and unable to find large displacements. In order to solve those problems, many strategies have been developed. We will present the method and analyze the benefits of a multi-layer strategy applied to Horn-Schunck.