Takumi Kaneko
Papers
1
Total Citations
3
H-Index
1
About
Takumi Kaneko is a researcher whose work sits at the intersection of computer vision, human activity recognition, and assistive robotics. His primary research focus involves developing computational methods for detecting human behaviors—particularly falls—using non-intrusive sensor systems. Kaneko’s most notable contribution is his work on human fall detection, where he pioneered the use of Cubic Higher-Order Local Auto-Correlation (CHLAC) features applied to skeletal image sequences captured by a Kinect sensor mounted on a mobile robot. This approach allows for the robust extraction of spatio-temporal geometric features from moving-image sequences, enabling the system to distinguish falls from other activities in real-time. Although his most-cited paper has garnered modest attention with 3 citations, its conceptual innovation lies in combining mobile robotics with advanced feature extraction for safety monitoring, a niche yet impactful area for elderly care and home automation. Kaneko’s work exemplifies how integrating low-cost depth sensors with sophisticated pattern recognition can pave the way for practical, autonomous health-assistive technologies.
Research Focus
Key Achievements
Top Papers
- 1Human fall detection using CHLAC features with skeletal image sequences3 citations · 2016