Phil Ammirato

University of North Carolina at Chapel Hill, University of North Carolina Health Care

Papers

Total Citations

241

H-Index

About

Phil Ammirato is a leading researcher in computer vision and robotics, whose work is pivotal in bridging the gap between perception and real-world robotic interaction. His primary research areas include active vision, object detection, and 3D pose estimation, with a strong focus on enabling robots to understand and manipulate objects in everyday environments. Ammirato’s most influential contribution is the creation of a large-scale dataset for active vision (193 citations), which provides over 20,000 RGB-D images and 50,000 bounding boxes across nine indoor scenes, serving as a benchmark for robotic vision tasks. He also advanced the field with “Fast Single Shot Detection and Pose Estimation” (25 citations), a method that simultaneously detects objects and estimates their 3D pose—critical for navigation and manipulation. Further notable work includes “Target Driven Instance Detection” (17 citations), which tailors detection for specific household objects, and “SymGAN” (6 citations), a novel approach to orientation estimation for symmetric objects without requiring manual annotation. Through these contributions, Ammirato has provided foundational tools and techniques that empower robots to perceive and interact with their surroundings more effectively, making his research essential for students and engineers developing next-generation autonomous systems.