Extension of Neural Disparity Estimation Networks for Reflective Regions through Specular Reflection Exploitation

Description

This thesis can be written in German and in English.

Disparity estimation plays a crucial role in computer vision, particularly for depth perception and 3D reconstruction. While neural networks have achieved remarkable performance in this domain, they struggle with smooth and textureless surfaces like plastics. In order to improve the performance of the neural network based approaches, a new disparity estimation algorithm shall be integrated into the state-of-the-art models.

Further information and the task description can be found in the following document.

Extension of Neural Disparity Estimation Networks for Reflective Regions through Specular Reflection Exploitation

Prerequisites

Advanced experience in programming with Python and good knowledge of image processing is required. Knowledge in Machine Learning/Deep Learning is helpful.

Category: Master Theses – Offered