Bernardo Teixeira

Papers

Total Citations

H-Index

About

Bernardo Teixeira is a researcher specializing in the intersection of deep learning and underwater robotics, with a primary focus on visual-based navigation for autonomous underwater vehicles (AUVs). His work addresses the critical challenge of enabling robust, persistent autonomy in the challenging underwater domain, where traditional visual odometry (VO) methods often fail due to poor lighting, turbidity, and feature-poor environments. Teixeira’s most cited paper, “Deep Learning for Underwater Visual Odometry Estimation” (2020, 37 citations), pioneers the application of deep neural networks to estimate camera motion in these difficult conditions, offering a data-driven alternative to conventional geometric approaches. His earlier work, “Deep Learning Approaches Assessment for Underwater Scene Understanding and Egomotion Estimation” (2019), systematically evaluates state-of-the-art deep learning architectures for both scene understanding and egomotion in confined underwater settings, providing a benchmark for the field. While deep learning methods still lag behind traditional algorithms in some metrics, Teixeira’s contributions are foundational, demonstrating the potential for neural networks to overcome the unique perceptual hurdles of underwater navigation. His research is pivotal for advancing the autonomy of AUVs in applications like ocean exploration, pipeline inspection, and environmental monitoring.