Papers
3
Total Citations
154
H-Index
2
About
Mejdi Dallel is a leading researcher at the intersection of virtual reality (VR), human–robot collaboration, and industrial artificial intelligence. His work focuses on solving one of the most pressing challenges in modern manufacturing: enabling robots to understand and predict human actions in real time. Dallel’s most impactful contribution is the development of the **InHARD (Industrial Human Action Recognition Dataset)**, a massive, real-world dataset comprising over 2 million frames from 16 subjects performing 13 distinct industrial actions. This dataset, cited over 60 times, provides the critical RGB+S data needed to train robust action recognition models for collaborative robotics. Building on this foundation, his highly cited 2022 paper (89 citations) introduces a groundbreaking method that uses VR to generate auto-labeled synthetic data for training these models, dramatically reducing the need for expensive manual annotation. By enabling more dynamic and responsive human-robot teams, Dallel’s work directly addresses the static nature of current collaboration, paving the way for safer, more efficient, and truly adaptive industrial workstations.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3