Motion estimation from noisy omnidirectional images
Brahim Alibouch, Mohammed Rziza, Amina Radgui, Driss Aboutajdine · 2014
Motion estimation methods in both perspective and omnidirectional cases presume that the input images are free-noise, and when using noisy images almost all of those methods yields poor results. In this paper we will examine the behavior and the performance of the phase-based method that we introduced in a previous work, in the presence of different kinds of noise.