Date of Award

12-2023

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

Committee Chair/Advisor

Dr. Feng Luo

Committee Member

Dr. Long Cheng

Committee Member

Dr. Nianyi Li

Abstract

As the field of computer vision continues to advance, the use of autonomous vehicles in military applications has increased and the tasks associated with these systems have grown in scope and complexity. These vehicles tend to operate in combat situations, presenting significant risk in the form of sensor damage. Since these autonomy algorithms rely on sensor information to navigate their environment, any threat to sensor functionality negatively impacts the reliability of the system. As a potential solution, I propose Dynamic Diffusion-based View Translation (DDVT), a novel computer vision algorithm capable of restoring image sensor function in ground vehicles through aerial to ground view translation using images from drones. Previous image to image techniques rely on well-defined and fixed translation parameters, however, the unpredictable nature of drone motion introduces uncertainty into the translation process as the translation distance between aerial and ground sensors is constantly changing. To address this challenge, DDVT leverages a transformer model backbone that introduces novel intermediate cross attention layers and translation label embeddings to facilitate dynamic view translation. DDVT is evaluated using numerous quantitative and visual benchmarks, ultimately confirming its potential for restoring sensor functionality through aerial to ground view translation using drones.

Author ORCID Identifier

0009-0003-1801-8209

Available for download on Tuesday, December 31, 2024

Share

COinS