Paper
17 October 2023 Sequentially trained, shallow neural networks for real-time 3D odometry
Author Affiliations +
Abstract
Fourier-domain correlation approaches have been successful in a variety of image comparison approaches but fail when the scenes, patterns, or objects in the images are distorted. Here, we utilize the sequential training of shallow neural networks on Fourier-preprocessed video to infer 3-D movement. The bio-inspired pipeline learns x, y, and z-direction movement from high-frame-rate, low-resolution, Fourier-domain preprocessed inputs (either cross power spectra or phase correlation data). Our pipeline leverages the high sensitivity of Fourier methods in a manner that is resilient to the parallax distortion of a forward-facing camera. Via sequential training over several path trajectories, models generalize to predict the 3-D movement in unseen trajectory environments. Models with no hidden layer are less accurate initially but converge faster with sequential training over different flightpaths. Our results show important considerations and trade-offs between input data preprocessing (compression) and model complexity (convergence).
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Frank Rodriguez, Baurzhan Muminov, and Luat T. Vuong "Sequentially trained, shallow neural networks for real-time 3D odometry", Proc. SPIE 12742, Artificial Intelligence for Security and Defence Applications, 127420X (17 October 2023); https://doi.org/10.1117/12.3005250
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Cameras

Fourier transforms

3D modeling

Data modeling

Video

3D image processing

Gyroscopes

RELATED CONTENT

UrbanScape
Proceedings of SPIE (May 01 2007)

Back to Top