Devin Guillory

Papers

Total Citations

H-Index

About

Devin Guillory is a researcher whose work sits at the intersection of computer vision, robotics, and cognitive science, with a particular focus on how machines can perceive and interact with dynamic, unstructured environments. His most cited paper, "A Probabilistic Framework for Real-time 3D Segmentation using Spatial, Temporal, and Semantic Cues" (2016, 46 citations), addresses a fundamental challenge in robotic perception: the need to segment a scene into discrete, trackable objects in real time. While conventional methods rely solely on spatial cues and fail under occlusion or motion, Guillory’s framework integrates temporal and semantic information into a probabilistic model, enabling robots to maintain coherent object identities even in cluttered, changing scenes. This contribution is critical for applications like autonomous navigation and manipulation, where robust object tracking is essential. Beyond this work, Guillory’s research spans human-robot interaction and the role of visual attention in learning, often bridging insights from human cognition to improve machine perception. His work is notable for its practical, real-time focus, making it directly applicable to deployed robotic systems. With a growing citation record and a commitment to making robots more perceptive and adaptive, Guillory is a rising voice in embodied AI and robotic vision.