1. What have the authors contributed in "A fast algorithm for vision-based hand gesture recognition for robot control" ?
The authors propose a fast algorithm for automatically recognizing a limited set of gestures from hand images for a robot control application.. The authors consider a fixed set of manual commands and a reasonably structured environment, and develop a simple, yet effective, procedure for gesture recognition.. Their approach contains steps for segmenting the hand region, locating the fingers, and finally classifying the gesture.. The authors demonstrate the effectiveness of the technique on real imagery.
read more
2. What is the motivation for this work?
What motivates us for this work is a robot navigation problem, in which the authors are interested in controlling a robot by hand pose signs given by a human.
read more
3. How many correct classifications have been obtained from the samples?
Out of 105 samples taken from 21 members of their laboratory, the authors have obtained 96 correct classifications which is approximately 91% of all images used in their experiments.
read more
4. How do the authors calculate the number of fingers in a gesture?
By counting the number of zero-to-one (black-to-white) transitions in this 1D signal, and subtracting one (for the wrist) leads to the estimated number of fingers active in the gesture.
read more
![Figure 3: Center of gravity and extreme points of the hand. Taken from [3].](/figures/figure-3-center-of-gravity-and-extreme-points-of-the-hand-310v14nu.webp)
![Figure 1: Set of hand gestures, or “counts” considered in our work. Adapted from [8].](/figures/figure-1-set-of-hand-gestures-or-counts-considered-in-our-12vf7v45.webp)