[CCoE Notice] Thesis Announcement: Omar Baig, "GPU Accelerated Tensor Voting"

[Greenwell, Stephen J]

[Thesis Defense Announcement at the Cullen College of Engineering]
GPU Accelerated Tensor Voting
Omar Baig

December 3, 2024, 12 p.m. - 1 p.m. (CST)

Location: Engineering Building 1 W309

Committee Chair:
David Mayerich Ph.D.

Committee Members:
Xin Fu, Ph.D. | Guoning Chen, Ph.D. | Badrinath Roysam, Ph.D.

Abstract

Gathering and extracting meaningful and reliable information from visual data
can be a challenging task, especially when dealing with noisy or incomplete data. Edge detection algorithms such as the Canny edge detector or the Sobel operator attempt to solve this issue by detecting and isolating the edges in images, but they often fail in the presence of noise or extracting higher-level structures. Noise removal techniques such as Gaussian blur exist, but they often come at a cost of losing important details in the image, leading to a smoothing effect that can obscure critical features such as edges, corners, or textures. This trade-off between noise reduction and feature retention limits their effectiveness in applications where both clarity and accuracy are crucial.

The Tensor Voting algorithm offers a robust solution for perceptual organization
and feature extraction by leveraging geometric and contextual information. However, the high computational complexity of the algorithm has limited its adoption in many cases. The research presented in this paper aims to describe the Tensor Voting algorithm, outline the math involved, and address the computational complexity of the algorithm by parallelzing the 2D Tensor Voting algorithm using CUDA on NVIDIA GPUs.

We discuss the challenges of parallezing the algorithm including a discussion
between the scatter and gather versions of the algorithm. Experimental results show that parallelizing the Tensor Voting algorithm with CUDA can achieve up to three orders of magnitude over the CPU implementation.
[Engineered For What's Next]

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://Bug.EGR.UH.EDU/pipermail/engi-dist/attachments/20241105/d99808d6/attachment-0001.html