Motion Detection in High Resolution Enhancement

Abstract

Shifted Superposition (SSPOS) is a resolution enhancement method
where apparent high-resolution content is displayed using a lowresolution
projection system with an opto-mechanical shifter. While
SSPOS-enhanced projectors have been showing promising results
in still images, they still suffer from motion artifacts in video contents.
Motivated by this, we present a novel approach to apparent
projector resolution enhancement for videos via motion-based
blurring module. We propose the use of a motion detection module
and a blurring module to compensate for both SSPOS-resulted and
natural motion artifacts in the video content. To accomplish this,
we combine both local and global motion estimation algorithms to
generate accurate dense flow fields. The detected motion regions
are enhanced using directional Gaussian filters. Preliminary results
show that the proposed method can produce accurate dense
motion vectors and significantly reduce the artifacts in videos.