Sign language recognition is vital for enhancing communication accessibility among the Deaf and hard-of-hearing communities.In Japan,approximately 360,000 individualswith hearing and speech disabilities rely on Japane...Sign language recognition is vital for enhancing communication accessibility among the Deaf and hard-of-hearing communities.In Japan,approximately 360,000 individualswith hearing and speech disabilities rely on Japanese Sign Language(JSL)for communication.However,existing JSL recognition systems have faced significant performance limitations due to inherent complexities.In response to these challenges,we present a novel JSL recognition system that employs a strategic fusion approach,combining joint skeleton-based handcrafted features and pixel-based deep learning features.Our system incorporates two distinct streams:the first stream extracts crucial handcrafted features,emphasizing the capture of hand and body movements within JSL gestures.Simultaneously,a deep learning-based transfer learning stream captures hierarchical representations of JSL gestures in the second stream.Then,we concatenated the critical information of the first stream and the hierarchy of the second stream features to produce the multiple levels of the fusion features,aiming to create a comprehensive representation of the JSL gestures.After reducing the dimensionality of the feature,a feature selection approach and a kernel-based support vector machine(SVM)were used for the classification.To assess the effectiveness of our approach,we conducted extensive experiments on our Lab JSL dataset and a publicly available Arabic sign language(ArSL)dataset.Our results unequivocally demonstrate that our fusion approach significantly enhances JSL recognition accuracy and robustness compared to individual feature sets or traditional recognition methods.展开更多
Mesh segmentation is one of the important issues in digital geometry processing. Region growing method has been proven to be a efficient method for 3D mesh segmentation. However, in mesh segmentation, feature line ext...Mesh segmentation is one of the important issues in digital geometry processing. Region growing method has been proven to be a efficient method for 3D mesh segmentation. However, in mesh segmentation, feature line extraction algorithm is computationally costly, and the over-segmentation problem still exists during region merging processing. In order to tackle these problems, a fast and efficient mesh segmentation method based on improved region growing is proposed in this paper. Firstly, the dihedral angle of each non-boundary edge is defined and computed simply, then the sharp edges are detected and feature lines are extracted. After region growing process is finished, an improved region merging method will be performed in two steps by considering some geometric criteria. The experiment results show the feature line extraction algorithm can obtain the same geometric information fast with less computational costs and the improved region merging method can solve over-segmentation well.展开更多
基金supported by the Competitive Research Fund of the University of Aizu,Japan.
文摘Sign language recognition is vital for enhancing communication accessibility among the Deaf and hard-of-hearing communities.In Japan,approximately 360,000 individualswith hearing and speech disabilities rely on Japanese Sign Language(JSL)for communication.However,existing JSL recognition systems have faced significant performance limitations due to inherent complexities.In response to these challenges,we present a novel JSL recognition system that employs a strategic fusion approach,combining joint skeleton-based handcrafted features and pixel-based deep learning features.Our system incorporates two distinct streams:the first stream extracts crucial handcrafted features,emphasizing the capture of hand and body movements within JSL gestures.Simultaneously,a deep learning-based transfer learning stream captures hierarchical representations of JSL gestures in the second stream.Then,we concatenated the critical information of the first stream and the hierarchy of the second stream features to produce the multiple levels of the fusion features,aiming to create a comprehensive representation of the JSL gestures.After reducing the dimensionality of the feature,a feature selection approach and a kernel-based support vector machine(SVM)were used for the classification.To assess the effectiveness of our approach,we conducted extensive experiments on our Lab JSL dataset and a publicly available Arabic sign language(ArSL)dataset.Our results unequivocally demonstrate that our fusion approach significantly enhances JSL recognition accuracy and robustness compared to individual feature sets or traditional recognition methods.
基金Supported by the National Natural Science Foundation of China(61272192,61379112)the NSFC-Guang dong Joint Fund(U1135003)
文摘Mesh segmentation is one of the important issues in digital geometry processing. Region growing method has been proven to be a efficient method for 3D mesh segmentation. However, in mesh segmentation, feature line extraction algorithm is computationally costly, and the over-segmentation problem still exists during region merging processing. In order to tackle these problems, a fast and efficient mesh segmentation method based on improved region growing is proposed in this paper. Firstly, the dihedral angle of each non-boundary edge is defined and computed simply, then the sharp edges are detected and feature lines are extracted. After region growing process is finished, an improved region merging method will be performed in two steps by considering some geometric criteria. The experiment results show the feature line extraction algorithm can obtain the same geometric information fast with less computational costs and the improved region merging method can solve over-segmentation well.
