Date of Award


Publication Type

Master Thesis

Degree Name



Electrical and Computer Engineering

First Advisor

Wu, Jonathan (Electrical and Computer Engineering)


Engineering, Electronics and Electrical.




In this thesis, three different depth map estimation techniques are presented. The first method uses SUSAN operator to detects the features, followed by an exponentially decaying function is employed to transfer the distance of the detected features by giving more weight to the nearer vicinity pixels of feature points, which helps to measure the clarity and depth of pixels. A robust, dual-tree complex wavelets and distance transformation based framework is developed for depth map estimation in second focus measure technique. The shift-invariance and better directionality of dual-tree complex wavelets helps to detects the features efficiently, which helps to estimate the depth of the scene more precisely. In third depth map estimation technique, focus measure is ensure by measuring local orientation energy using a quadrature pair of steerable filters of the detected features. The experiments and results validates the effectiveness of proposed feature based depth map estimation approach.