Motion estimation based on optical flow and an artificial neural network (ANN) |
| |
Authors: | Jiafeng Zhang Feifei Zhang Masanori Ito |
| |
Affiliation: | 1. Department of Applied Ocean Engineering, Tokyo University of Marine Science and Technology, 2-1-6 Etchujima, Koto-Ku, Tokyo, Japan
|
| |
Abstract: | Motion estimation provides an attractive alternative to cumbersome interface devices for human-computer interaction (HCI).
Worthy of note is that the visual recognition of hand gestures can help to achieve an easy and natural interaction between
human and computer. The interfaces of HCI and other virtual reality systems depend on accurate, real-time hand and fingertip
tracking for an association between real objects and the corresponding digital information. However, they are expensive, and
complicated operations can make them troublesome. We are developing a real-time, view-based gesture recognition system. The
optical flow is estimated and segmented into motion fragments. Using an artificial neural network (ANN), the system can compute
and estimate the motions of gestures. Compared with traditional approaches, theoretical and experimental results show that
this method has simpler hardware and algorithms, but is more effective. It can be used in moving object recognition systems
for understanding human body languages. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|