Max Robotics — Global Service Robot Directory + US Certification

About

Michel Antunes is a computer vision researcher whose work focuses on 3D reconstruction, stereo vision, and odometry for robotics. His most influential contribution is a plane-based odometry method using RGB-D cameras (45 citations), which estimates sensor motion from planar surfaces in successive frames—a critical capability for robot navigation and mapping. Antunes has also investigated whether stereo vision can replace more expensive Laser Rangefinders, exploring cost-effective sensing for robotic systems. His research extends to dense stereo matching and 3D reconstruction, including parallel refinement of slanted 3D models using symmetry-induced stereo, and distributed processing for efficient depth estimation. He developed stereo methods for estimating depth along virtual cut planes, enabling selective scene analysis. Antunes’ work on accelerating SymStereo—a symmetry-based stereo algorithm—demonstrates his interest in computationally efficient, parallel processing for real-time applications. With a portfolio spanning from foundational odometry to advanced 3D reconstruction, his contributions have practical implications for autonomous vehicles, robotics, and intelligent systems, making him a notable figure in applied computer vision.