Takeshi Onishi
Papers
1
Total Citations
34
H-Index
1
About
Takeshi Onishi is a leading researcher in computer vision and robotics, with a focus on multi-task learning for real-world perception systems. His most cited work, "MT-DSSD: Deconvolutional Single Shot Detector Using Multi Task Learning for Object Detection, Segmentation, and Grasping Detection" (2020, 34 citations), introduces a groundbreaking unified network that simultaneously performs object detection, semantic segmentation, and suction-based grasping detection. This innovation streamlines robotic manipulation by enabling a single model to understand and interact with objects, reducing computational overhead and improving efficiency in autonomous systems. Onishi’s contributions bridge the gap between visual perception and physical action, advancing applications in industrial automation and service robotics. His work has garnered attention for its practical impact, with the MT-DSSD framework influencing subsequent research in multi-task architectures for embodied AI. By integrating deconvolutional layers and shared feature representations, Onishi demonstrates how deep learning can solve complex, multi-objective problems in real time. His research continues to inspire students and engineers exploring the intersection of computer vision, deep learning, and robotics, making him a notable figure in the development of intelligent, task-aware systems.
Research Focus
Key Achievements
Top Papers
- 1