log in  |  register  |  feedback?  |  help  |  web accessibility
PhD Proposal: Towards Leveraging Sparsity for Immersive and Interactive 3D Displays
Susmija Jabbireddy
Wednesday, November 2, 2022, 12:00-2:00 pm Calendar
  • You are subscribed to this talk through .
  • You are watching this talk through .
  • You are subscribed to this talk. (unsubscribe, watch)
  • You are watching this talk. (unwatch, subscribe)
  • You are not subscribed to this talk. (watch, subscribe)
Augmented and virtual reality (VR and AR) are bridging the gap between the physical and the virtual worlds. The ultimate goal of AR and VR technology is to present three-dimensional (3D) images at high frame rates for a realistic, immersive, and interactive viewing experience. However, as the resolution of the AR and VR devices increases, the computational intensity increases, placing a burden on the time and energy resources to render, stream, and display information. This research explores methods that leverage sparsity to enable effective augmented and virtual environments.

First, we discuss various foveated rendering techniques. With the advent of real-time eye tracking systems and an increase in the resolution and field of view in modern AR and VR headsets, foveated rendering becomes crucial to achieve real-time rendering. We review the current state of the field and provide a taxonomy of various foveation techniques that can be used as a guide for developing foveated rendering methods.

Then we explore methods to improve the quality of images from sparse Monte Carlo samples in volumetric rendering. Monte Carlo path tracing can provide stunning visualization of volumetric data. In the medical domain, photo-realistic visualizations improve the perception of internal structures for time-critical medical procedures as well as training for medical professionals. However, the number of samples to be computed is extremely large to achieve noise-free images. We show how deep-learning based denoising techniques can be integrated with Monte Carlo volumetric rendering to achieve high-quality images of volumetric data at interactive rates.

Next, we present our research towards developing energy-efficient holographic displays. Holographic displays are considered true 3D displays, with the potential to emulate all the depth cues of human vision. Nanophotonic phased array (NPA) is a novel emerging technology for holographic displays. NPAs offer the unique advantages of compact chip-sized displays and very high refresh rates. However, to produce a high-resolution wide field-of-view holographic image, building a large-scale NPA is limited by the significant power consumption and circuit complexity. We present algorithms to generate sparse holographic patterns and show that we can produce high-resolution holographic images with high quality using as few as 10% of the array pixels.

For future research, we will develop techniques to generate sparse NPA configurations for 3D holographic images. In addition, we will attempt to improve the computational efficiency of generating sparse holograms.
Examining Committee


Dr. Amitabh Varshney

Department Representative:

Dr. Bahar Asgari


Dr. Matthias Zwicker

Dr. Martin Peckerar


Dr. Mario Dagenais


Susmija is pursuing a Ph.D. in Computer Science at the University of Maryland, College Park, advised by Prof. Amitabh Varshney. Susmija's research interests involve computer graphics, 3D displays, and computer vision, focusing on solving problems to develop immersive and interactive visual content. She received her master's and bachelor's degree in Computer Science from the Indian Institute of Technology, Kharagpur.

This talk is organized by Tom Hurst