Towards Designing Robust Deep Learning Models for 3D Understanding

Abstract

Deep Neural Networks (DNNs)  have shown huge success over the years to solve many 2D computer vision tasks driven by massive labeled 2D datasets and advancements in 2D vision models, but less success is witnessed on 3D vision tasks. This dissertation proposes innovative approaches to enhance the robustness of DNNs for 3D understanding and in 3D settings. The research focuses on two main areas: adversarial robustness on 3D data and setups, and the robustness of DNNs to realistic 3D scenarios. Two paradigms for 3D understanding are discussed: representing 3D as a set of 3D points and performing 2D processing of multiple images of the 3D data.

Brief Biography

Abdullah is a Ph.D. candidate in Electrical and Computer Engineering in Prof. Bernard Ghanem’s Computer Vision group at KAUST. He focuses on developing robust 3D computer vision techniques and tackling the adversarial robustness of deep neural networks. Abdullah has won a “best paper award” at the European Conference of Computer Vision ECCV 2020 and the NEOM AI Challenge 2020 and is a recipient of the Ibn Rushd postdoc fellowship award 2022, among other national and international distinctions (AAI 2020, ECCV2020, ICCV 2021, and ICLR 2023).  Abdullah is also the founder and president of fihm.ai, the largest Arabic online platform dedicated to teaching, educating, and spreading awareness about AI and deep learning technologies and applications.