Taylor Denouden

Papers

Total Citations

H-Index

About

Taylor Denouden is a researcher whose work lies at the critical intersection of computer vision and safety-critical autonomous systems, with a primary focus on uncertainty quantification in object detection. Denouden’s most influential contribution, the 2018 paper "Calibrating Uncertainties in Object Localization Task" (19 citations), addresses a fundamental challenge in deploying deep learning models for real-world applications like autonomous driving and surgical robotics: the need for reliable, calibrated prediction uncertainties. This work pioneered methods to ensure that object detection modules not only identify objects but also accurately estimate the probability of each prediction, enabling safer decision-making. By tackling the gap between raw model outputs and trustworthy probabilistic estimates, Denouden’s research directly supports the development of robust perception systems where overconfident errors could have catastrophic consequences. While still early in their career, Denouden’s contributions are already recognized as foundational steps toward bridging theoretical uncertainty calibration with practical deployment in high-stakes environments. Their work continues to influence researchers and engineers striving to make autonomous systems not just intelligent, but demonstrably trustworthy.