Papers
6
Total Citations
290
H-Index
4
About
Alina Roitberg is a researcher whose work bridges human-robot interaction, computer vision, and neuroscience-inspired computing. Her research focuses primarily on human activity recognition, particularly in industrial and collaborative robotics contexts, where understanding human behavior is essential for safe and intuitive human-robot cooperation. Roitberg's most influential contributions include pioneering multimodal and skeleton-based approaches to activity recognition in industrial robotic workcells, work that has collectively garnered over 150 citations and helped establish foundational methods for interpreting human actions in real-world manufacturing environments. Her early papers demonstrated how spatial and temporal skeletal features could enable robots to understand worker activities with practical reliability. She also contributed to the ambitious Neurorobotics Platform project (117 citations), which connected biologically realistic spiking neural network brain models to robotic embodiments within rich simulation environments — a landmark interdisciplinary effort spanning neuroscience, biology, and computer science. More recently, Roitberg has tackled emerging challenges such as learning from noisy annotations in action recognition and leveraging synthetic data to improve model generalizability in privacy-sensitive household settings. Her trajectory reflects a sustained commitment to making robots more perceptive, adaptive, and safe partners for humans across diverse real-world applications.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4Skeleton-Based Human Action Recognition with Noisy Labels11 citations · 2024
- 5
- 6