Hand Detection Using Cascade of Softmax Classifiers

Sliding-window based multiclass hand posture detections are often performed by detecting postures of each predefined category using an independent detector, which makes it lack efficiency and results in high postures confusion rates in real-time applications. To tackle such problems, in this work, a...

Full description

Saved in:
Bibliographic Details
Main Authors: Yan-Guo Zhao, Feng Zheng, Zhan Song
Format: Article
Language:English
Published: Wiley 2018-01-01
Series:Advances in Multimedia
Online Access:http://dx.doi.org/10.1155/2018/9204854
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832563796217954304
author Yan-Guo Zhao
Feng Zheng
Zhan Song
author_facet Yan-Guo Zhao
Feng Zheng
Zhan Song
author_sort Yan-Guo Zhao
collection DOAJ
description Sliding-window based multiclass hand posture detections are often performed by detecting postures of each predefined category using an independent detector, which makes it lack efficiency and results in high postures confusion rates in real-time applications. To tackle such problems, in this work, an efficient cascade detector that integrates multiple softmax-based binary (SftB) models and a softmax-based multiclass (SftM) model is investigated to perform multiclass posture detection in parallel. The SftB models are used to distinguish the predefined postures from the background regions, and the SftM model is applied to discriminate among all the predefined hand posture categories. Another usage of the cascade structure is that it could effectively decompose the complexity of background pattern space and therefore improve the detection accuracy. In addition, to balance the detection accuracy and efficiency, the HOG features of increasing resolutions will be adopted by classifiers of increasing stage-levels in the cascade structure. The experiments are implemented under various scenarios with complicated background and challenging lightings. Results show the superiority of the proposed SftB classifiers over the traditional binary classifiers such as logistic regression, as well as the accuracy and efficiency improvements brought by the softmax-based cascade architecture compared with the noncascade multiclass softmax detectors.
format Article
id doaj-art-a47b7da09a214453bca989bcc78e46cb
institution Kabale University
issn 1687-5680
1687-5699
language English
publishDate 2018-01-01
publisher Wiley
record_format Article
series Advances in Multimedia
spelling doaj-art-a47b7da09a214453bca989bcc78e46cb2025-02-03T01:12:31ZengWileyAdvances in Multimedia1687-56801687-56992018-01-01201810.1155/2018/92048549204854Hand Detection Using Cascade of Softmax ClassifiersYan-Guo Zhao0Feng Zheng1Zhan Song2Shenzhen Key Laboratory of Virtual Reality and Human Interaction Technology, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, ChinaSwanson School of Engineering, The University of Pittsburgh, Pittsburgh, PA 15261, USAShenzhen Key Laboratory of Virtual Reality and Human Interaction Technology, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, ChinaSliding-window based multiclass hand posture detections are often performed by detecting postures of each predefined category using an independent detector, which makes it lack efficiency and results in high postures confusion rates in real-time applications. To tackle such problems, in this work, an efficient cascade detector that integrates multiple softmax-based binary (SftB) models and a softmax-based multiclass (SftM) model is investigated to perform multiclass posture detection in parallel. The SftB models are used to distinguish the predefined postures from the background regions, and the SftM model is applied to discriminate among all the predefined hand posture categories. Another usage of the cascade structure is that it could effectively decompose the complexity of background pattern space and therefore improve the detection accuracy. In addition, to balance the detection accuracy and efficiency, the HOG features of increasing resolutions will be adopted by classifiers of increasing stage-levels in the cascade structure. The experiments are implemented under various scenarios with complicated background and challenging lightings. Results show the superiority of the proposed SftB classifiers over the traditional binary classifiers such as logistic regression, as well as the accuracy and efficiency improvements brought by the softmax-based cascade architecture compared with the noncascade multiclass softmax detectors.http://dx.doi.org/10.1155/2018/9204854
spellingShingle Yan-Guo Zhao
Feng Zheng
Zhan Song
Hand Detection Using Cascade of Softmax Classifiers
Advances in Multimedia
title Hand Detection Using Cascade of Softmax Classifiers
title_full Hand Detection Using Cascade of Softmax Classifiers
title_fullStr Hand Detection Using Cascade of Softmax Classifiers
title_full_unstemmed Hand Detection Using Cascade of Softmax Classifiers
title_short Hand Detection Using Cascade of Softmax Classifiers
title_sort hand detection using cascade of softmax classifiers
url http://dx.doi.org/10.1155/2018/9204854
work_keys_str_mv AT yanguozhao handdetectionusingcascadeofsoftmaxclassifiers
AT fengzheng handdetectionusingcascadeofsoftmaxclassifiers
AT zhansong handdetectionusingcascadeofsoftmaxclassifiers