Ahmad AlMughrabi

I am a predoctoral scholar at the Universitat de Barcelona specialising in Deep Learning, Computer Graphics, and Computer Vision in Barcelona, Spain.

Moreover, I am an experienced Software Engineer with a demonstrated history of working in the computer software industry. I am skilled in Java, Python, FE & DevOps technologies. Energetic engineering professional with an Honours Master's degree focused in Computer Science from the University of Jordan in 2020, Amman, Jordan. He received an Honours BSc in Computer Science from the Zarqa University in 2012 in Zarqa, Jordan.

Email  /  CV  /  Bio  /  Google Scholar  /  Linkedin  /  Github

profile photo
Research

I'm interested in computer vision, deep learning, and computer graphics. My supervisors are:

Professor Petia Radeva is a Full Professor with the Universitat de Barcelona (UB), Barcelona, Spain, the Head of the Consolidated Research Group “Computer Vision and Machine Learning” with the Computer Vision and Machine Learning at the University of Barcelona (CVMLUB), UB, and a Senior Researcher with Computer Vision Center, Barcelona, Spain.
Professor Ricardo Marques, a Serra HĂșnter Lecturer at the Department of Mathematics and Informatics of the Universitat de Barcelona (UB), Barcelona, Spain.
Representative papers are highlighted.

FoodMem: Near Real-time and Precise Food Video Segmentation

Ahmad AlMughrabi AdriĂĄn GalĂĄn Ricardo Marques Petia Radeva

project page / arXiv

FoodMem is a novel framework designed to segment food items from video sequences of 360-degree unbounded scenes. FoodMem can consistently generate masks of food portions in a video sequence, overcoming the limitations of existing semantic segmentation models, such as flickering and prohibitive inference speeds in video processing contexts.

Pre-NeRF 360: Enriching Unbounded Appearances for Neural Radiance Fields

Ahmad AlMughrabi Umair Haroon Ricardo Marques* Petia Radeva*

project page / arXiv

Our solution overcomes several obstacles that plagued earlier versions of NeRF, including handling multiple video inputs, selecting keyframes, and extracting poses from real-world frames that are ambiguous and symmetrical.

MomentsNeRF: Incorporating Orthogonal Moments in Convolutional Neural Networks for One or Few-Shot Neural Rendering

Ahmad AlMughrabi Ricardo Marques Petia Radeva

project page / arXiv

Our solution overcomes several obstacles that plagued earlier versions of NeRF in one and few-shot neural rendering, including handling artefacts, reducing the learning complexity, and eliminating noise.