Kenji Maseمشاهده پروفایل
پژوهشگر
Kenji Mase is a prolific researcher in computer vision and human-computer interaction, with publications spanning over three decades. His recent work focuses on developing advanced deep learning architectures for fine-grained recognition tasks, particularly in transportation safety applications like distracted driver detection. His research explores multimodal sensing approaches, including novel garment-based IMU systems for human activity recognition. Recent publications demonstrate innovations in attention mechanisms and lightweight neural network designs suitable for edge computing applications. Mase's work consistently addresses practical implementation challenges through model compression techniques like knowledge distillation and neural architecture search. This research direction bridges theoretical machine learning with real-world deployment constraints.






