Shailza Sharma

Shailza Sharma

Profile

I am a Research Fellow working on the DEEPVOLC project at the University of Leeds, where I develop deep learning models to forecast volcanic activity using spatio-temporal satellite data. My expertise spans computer vision, generative modeling, deep learning, and generative AI.

I hold a Ph.D. in Semantic and Depth-Aware Generative Adversarial Network for Image Super-Resolution.

Previously, as a Research Scientist at Kroop AI, I worked on a pioneering project to achieve lip-syncing in talking face videos of arbitrary identities to match target speech segments, utilizing advanced techniques in generative adversarial networks (GANs) and audio-visual alignment.

My research interests include generative modeling, convolutional neural networks (CNNs), GANs, super-resolution, deep fake analysis, and face hallucination. I am passionate about leveraging AI to address real-world challenges in diverse domains, from natural phenomena to human-centric applications.

Responsibilities

  • Developing deep learning models for the DEEPVOLC project to forecast volcanic activity using spatio-
  • temporal satellite data. Responsibilities include designing data pipelines, implementing advanced
  • deep learning based architectures, and conducting large-scale training and testing on HPC systems.
<h4>Research projects</h4> <p>Some research projects I'm currently working on, or have worked on, will be listed below. Our list of all <a href="https://environment.leeds.ac.uk/dir/research-projects">research projects</a> allows you to view and search the full list of projects in the faculty.</p>