King Abdullah University of Science and Technology
Bibi graduated with a B.Sc. in Electrical Engineering from Kuwait University with class honors in 2014. He received his M.Sc. in Electrical Engineering at King Abdullah University of Science and Technology (KAUST) in 2016 with class honors. He rolled over to the PhD program in the same year where he has since been a PhD student and a research assistant. His research interest span the field of computer vision, machine learning and large scale optimization and has contributed to more than 12 publications in the past 5 years in top tier conferences (CVPR/ICCV/ECCV). Some of his papers were selected for full oral and spotlight presentations at CVPR and ECCV. In 2018, he won the best KAUST poster award on new tensor factorization strategy for convolutional sparse coding and was recognized as an outstanding reviewer for CVPR18. He was a visiting PhD research student for 6 months at Intel Labs in Munich, Germany in 2018.
Recently, he has started working on analyzing a block of layers in deep neural networks from probabilistic, tropical geometric, bound, and optimization perspectives.
A basket of computer vision research problems: Automation, Autonomous navigation, Video Understanding and Theory.
introduce the driving force behind the recent successes and developments in computer vision and machine learning, often referred to as artificial intelligence. In particular, I will introduce convolutional neural networks and the basic principals of the mechanistics that make them work. I will also highlight the key differences between ConvNet and classical approaches.
The second part of the talk will be dedicated to various applications and trends in computer vision. In particular, the learning systems and models developed at our group IVUL (Image and Video Understanding Lab) in KAUST. The applications will range from automation to autonomous navigation and video understanding. Applications such as visual object tracking, activity recognition and detection in video, teaching drones to race and autonomous driving in simulation. I will also briefly talk about training robots and agents in simulation through a reward mechanism where such a learning approach is often referred to as reinforcement learning. Lastly, I will end the talk with some more theoretical aspects of computer vision and the open research problems.