Pointing gesture recognition using compressed sensing for training data reduction

Authors:
Masahiro Iwasaki;Kaori Fujinami
Affiliations:
Tokyo University of Agriculture and Technology, Tokyo, Japan;Tokyo University of Agriculture and Technology, Tokyo, Japan
Venue:
Proceedings of the 2013 ACM conference on Pervasive and ubiquitous computing adjunct publication
Year:
2013

Citing 8
Cited 0

An HMM-Based Threshold Model Approach for Gesture Recognition

IEEE Transactions on Pattern Analysis and Machine Intelligence
Database-friendly random projections

PODS '01 Proceedings of the twentieth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
Random projection in dimensionality reduction: applications to image and text data

Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining
Hidden Markov Model Based Continuous Online Gesture Recognition

ICPR '98 Proceedings of the 14th International Conference on Pattern Recognition-Volume 2 - Volume 2
Gestures are strings: efficient online gesture spotting and classification using string matching

Proceedings of the ICST 2nd international conference on Body area networks
uWave: Accelerometer-based personalized gesture recognition and its applications

Pervasive and Mobile Computing
Continuous Malayalam speech recognition using Hidden Markov Models

Proceedings of the 1st Amrita ACM-W Celebration on Women in Computing in India
A Novel Accelerometer-Based Gesture Recognition System

IEEE Transactions on Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we investigate training data reduction for the pointing gesture recognition with compressed sensing. The pointing gesture is one of activities during pointing and calling that is carried out by workers to keep occupational safety and correctness. Compressed sensing is used for gesture recognition and considered the impacts of the gesture duration difference among user. However, the different force among users may affect to the recognition. As a result of the experiment, F-measure is improved 0.18 compared with the DTW even only the data obtained from others is used. Moreover, we found that the user-dependency varies for each subject. Therefore, we tested to recognize the pointing gestures of all subjects by using the training data of only specific users. The test showed that the recognition model with training data from 4 specific subjects provided the same accuracy as the one from 11 subjects. This result suggested the feasibility of reduction for subjects who need to acquire the training data.