Luca Minciullo
Papers
3
Total Citations
78
H-Index
3
About
Luca Minciullo is a leading researcher in computer vision and robotics, specializing in 6D object pose and shape estimation, as well as robotic grasping. His work addresses a critical bottleneck in real-world robotics: enabling machines to interact with hundreds of unseen objects, not just a handful of known instances. Minciullo’s major contributions include the development of CPS and CPS++, pioneering deep learning frameworks for class-level 6D pose and shape estimation from monocular images. These methods allow robots to infer the full 3D pose and shape of novel objects in a category, moving beyond instance-specific approaches. His work on DemoGrasp introduces few-shot learning for robotic grasping, where a robot can learn to grasp new objects from just a single human demonstration, dramatically improving adaptability. With over 78 citations across his top papers, Minciullo’s research has significant impact, bridging the gap between controlled lab settings and dynamic, real-world environments. His achievements are foundational for the next generation of autonomous robots that can seamlessly integrate into everyday human spaces.
Research Focus
Key Achievements
Top Papers
- 1
- 2DemoGrasp: Few-Shot Learning for Robotic Grasping with Human Demonstration29 citations · 2021
- 3CPS: Class-level 6D Pose and Shape Estimation From Monocular Images14 citations · 2020