M. Umair Mukati

I am an Early Stage Researcher in Marie Curie RealVision ITN, where I am conducting research broadly on the topic of light field coding and processing.

I am a Ph.D. candidate at the Department of Photonics Engineering in Technical University of Denmark and a part of Coding and Visual Communication group headed by Prof. Søren Forchhammer. I received my B.E. degree in Industrial Electronics Engineering from NED University of Engineering and Technology, Karachi, Pakistan and M.S. degree in Electrical-Electronics Engineering from Istanbul Medipol University, Istanbul, Turkey in 2014 and 2017, respectively.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  Github

I'm interested in 3D computer vision, data compression and machine learning.

A novel learning-based approach utilizing attention mechanism to synthesize novel views of a light-field image using a sparse set of input views.

An improved scheme for distributed light field coding involving deep-learning for view-synthesis and uncertainty prediction.

An efficient lossless and near-lossless compression scheme for light field images by improving prediction and context formation process in EPIC.

A distributed compression scheme for light field to achieve a low complexity encoder leveraging deep-learning-based view synthesis for side information generation.

An efficient lossless compression scheme for light field images based on CALIC.

A light field super-resolution approach by applying microshift in the light field sensor position through motorized translation stage.

Improving light field's spatial resolution by utilizing light field sensor and traditional photo-sensor utilizing a single optical lens and a beam splitter.

Synthetic aperture enhancement by stitching several light field captures using multiple view geometry.