3D Face Reconstruction from Front and Profile Image

Date of Award

2021

Degree Name

M.C.S. (Master of Computer Science)

Department

Department of Computer Science

Advisor/Chair

Mehdi R. Zargham

Abstract

Three dimensionality (3D) face modeling is an advanced and challenging feature for computer vision, and our goal is to implement it using various methods to bring 3D models closer to reality. Although many algorithms for construction of 3D model from two dimensional (2D) images are present, we propose a new approach using front and profile images with various image processing techniques for small computing devices. Basic methods such as resizing, denoise, overlay, blending etc. will be used for generation of the UV-map of texture, but as its core element, it relies on the Haar Cascade face detection algorithm. For structure or mesh, a shape detector with 68 landmarks to identify the shape of the face in the image and compare it with our own dataset for most similar structure. Though we have achieved good results from the proposed approach, there is potential to improve by making the model an identical replica.

Keywords

2D to 3D, Haar Cascade Classifier, OpenCV, dlib, UV-map and Mesh

Rights Statement

Copyright © 2021, author

Link to Full Text

Share

COinS