Realtime estimation of human hand posture for robot hand control

Abstract

The authors propose a system for searching the shape of human hands and fingers in real time and with high accuracy, without using any special peripheral equipment such as range sensor, PC cluster, etc., by a method of retrieving similar image quickly with high accuracy from a large volume of image database containing the complicated shapes and selfocclusions. In designing the system, we constructed a database in a way to be adaptable even to differences among individuals, and searched CG images of hand similar to unknown hand image, through extraction of characteristics using high-order local auto-correlational patterns, reduction of the amount of characteristics centering on principal component analysis, and prior rearrange-ment of data corresponding to the amount of characteristics. As a result of experiments, our system performed high-accuracy estimation of human hand shape where mean error was 7 degrees in finger joint angles, with the processing speed of 30 fps or over.

Fast finger tracking system for in-air typing interface

“…3D shape reconstruction using multiple cameras [6] could obtain detailed hand shape, but it is ill-suited to portable devices in terms of size. Some hand gesture recognition system using a single camera use hand pose database [7,8], which requires large memory and it is difficult to implement in a portable device. Vision-based hand gesture recognition often uses a silhouette image of hand, which wouldn't work well in the presence of cluttered background.…”