This inventor holds 1 USPTO granted patent. Top assignees: Astemo Americas, Inc., University of Michigan. Active years: 2026.
Company Filing History:
1 out of 1 
1 out of 3,511 
Years Active: 2026
Title: Arima Hidetoshi: Innovator in Object Detection Technology
Introduction
Arima Hidetoshi is a notable inventor based in Farmington Hills, MI (US). He has made significant contributions to the field of object detection through his innovative patent. His work focuses on enhancing the efficiency of object detection models, which are crucial in various applications.
Latest Patents
Hidetoshi holds a patent titled "Adaptive intelligence for object detection via dynamic model composition." This patent addresses the limitations of current object detection models, which require extensive training cycles on powerful hardware. His invention proposes a method to interchange the components of object detection models without the need for retraining. This approach allows for the rapid construction and evaluation of new model designs. By introducing a thin adapter, Hidetoshi enables the recomposition of pre-trained sub-models, resulting in a suite of new models that achieve distinct accuracy and compute profiles at minimal training costs. These new models have demonstrated improvements of up to 3% in mean Average Precision (mAP) on the COCO dataset while reducing training costs by up to 99% compared to traditional methods.
Career Highlights
Throughout his career, Hidetoshi has worked with esteemed organizations, including the University of Michigan and Astemo Americas, Inc. His experience in these institutions has contributed to his expertise in the field of object detection and model optimization.
Collaborations
Hidetoshi has collaborated with notable colleagues such as Jason Mars and Zongyu Chen. Their combined efforts have further advanced the research and development of innovative technologies in object detection.
Conclusion
Arima Hidetoshi's contributions to the field of object detection through his patent and collaborative efforts highlight his role as an influential inventor. His innovative approach to model composition is paving the way for more efficient and effective object detection technologies.