Waleed H. Abdulla, Felix Marattukalam, Vedrana Krivokuća Hahn
{"title":"Exploring Human Biometrics: A Focus on Security Concerns and Deep Neural Networks","authors":"Waleed H. Abdulla, Felix Marattukalam, Vedrana Krivokuća Hahn","doi":"10.1561/116.00000021","DOIUrl":"https://doi.org/10.1561/116.00000021","url":null,"abstract":"","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67080034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Dual-branch Convolutional Network Architecture Processing on both Frequency and Time Domain for Single-channel Speech Enhancement","authors":"Kanghao Zhang, Shulin He, Hao Li, Xueliang Zhang","doi":"10.1561/116.00000083","DOIUrl":"https://doi.org/10.1561/116.00000083","url":null,"abstract":"","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67081300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Order learning aims to learn the ordering relationship among objects by comparing them. Recently, several order learning techniques have achieved great performances on various computer vision tasks. In this paper, we provide an overview of these order learning techniques. First, we briefly discuss conventional rank estimation algorithms related to order learning. Second, we review the order learning techniques in detail. Third, we discuss the results of order learning on three vision applications: facial age estimation, historical color image (HCI) classification, and aesthetic quality assessment.
{"title":"Order Learning – An Overview","authors":"Seon-Ho Lee, Nyeong-Ho Shin, Chang-Su Kim","doi":"10.1561/116.00000226","DOIUrl":"https://doi.org/10.1561/116.00000226","url":null,"abstract":"Order learning aims to learn the ordering relationship among objects by comparing them. Recently, several order learning techniques have achieved great performances on various computer vision tasks. In this paper, we provide an overview of these order learning techniques. First, we briefly discuss conventional rank estimation algorithms related to order learning. Second, we review the order learning techniques in detail. Third, we discuss the results of order learning on three vision applications: facial age estimation, historical color image (HCI) classification, and aesthetic quality assessment.","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67081746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Optical Flow Regularization of Implicit Neural Representations for Video Frame Interpolation","authors":"Weihao Zhuang, Tristan Hascoet, Xunquan Chen, Ryoichi Takashima, Tetsuya Takiguchi","doi":"10.1561/116.00000218","DOIUrl":"https://doi.org/10.1561/116.00000218","url":null,"abstract":"","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135446724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lufeng Chen, Hongqin Xie, Zicheng Liu, Bin Li, Hong Cheng
The aging population and increased number of individuals with motor dysfunction pose significant challenges to the workforce. This situation is further exacerbated by a declining working-age population, which has resulted in labor shortages. A potential remedy to these issues lies in the employment of wearable robots. As a form of human-robot collaboration, these devices can augment motor capabilities and offer assistance with various motor functions. To this end, this paper presents a systematic review of the current research status of wearable robots, focusing on the applications of Supernumerary Robotic Limbs (SRL) and exoskeletons for task assistance and motor function restoration in the field of industrial and rehabilitation, respectively. The paper also deliberates on the research trends, challenges, and prospective directions of human-robot interaction and control strategies regarding wearable robots.
{"title":"Exploring Challenges and Opportunities of Wearable Robots: A Comprehensive Review of Design, Human-Robot Interaction and Control Strategy","authors":"Lufeng Chen, Hongqin Xie, Zicheng Liu, Bin Li, Hong Cheng","doi":"10.1561/116.00000156","DOIUrl":"https://doi.org/10.1561/116.00000156","url":null,"abstract":"The aging population and increased number of individuals with motor dysfunction pose significant challenges to the workforce. This situation is further exacerbated by a declining working-age population, which has resulted in labor shortages. A potential remedy to these issues lies in the employment of wearable robots. As a form of human-robot collaboration, these devices can augment motor capabilities and offer assistance with various motor functions. To this end, this paper presents a systematic review of the current research status of wearable robots, focusing on the applications of Supernumerary Robotic Limbs (SRL) and exoskeletons for task assistance and motor function restoration in the field of industrial and rehabilitation, respectively. The paper also deliberates on the research trends, challenges, and prospective directions of human-robot interaction and control strategies regarding wearable robots.","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135659597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Reversible Data Hiding in Compressible Encrypted Images with Capacity Enhancement","authors":"Ryota Motomura, Shoko Imaizumi, H. Kiya","doi":"10.1561/116.00000014","DOIUrl":"https://doi.org/10.1561/116.00000014","url":null,"abstract":"","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67079973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"EMS2L: Enhanced Multi-Task Self-Supervised Learning for 3D Skeleton Representation Learning","authors":"Lilang Lin, Jiaying Liu","doi":"10.1561/116.00000022","DOIUrl":"https://doi.org/10.1561/116.00000022","url":null,"abstract":"","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67080045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
R. Jinzai, K. Yamaoka, S. Makino, Nobutaka Ono, T. Yamada, M. Matsumoto
Because inter–channel time differences (ICTDs) between signals detected by real microphones mounted close to each other are much smaller than inter–aural time differences (ITDs) for sound image localization, sound images are localized at azimuths different from those of sound sources. In this paper, we propose a virtual microphone technique, which simulates binaural signals by equalizing ICTDs to ITDs, to localize sound images at azimuths of the sound sources with reference to the real microphones. Binaural signals simulated by the proposed method were examined objectively and subjectively by tests on two-sound-image localization. The tests revealed that the two sound images were localized at azimuths of the sound sources with reference to the real microphones.
{"title":"Virtual Microphone Technique for Binauralization for Multiple Sound Images on 2–Channel Stereo Signals Detected by Microphones Mounted Closely","authors":"R. Jinzai, K. Yamaoka, S. Makino, Nobutaka Ono, T. Yamada, M. Matsumoto","doi":"10.1561/116.00000079","DOIUrl":"https://doi.org/10.1561/116.00000079","url":null,"abstract":"Because inter–channel time differences (ICTDs) between signals detected by real microphones mounted close to each other are much smaller than inter–aural time differences (ITDs) for sound image localization, sound images are localized at azimuths different from those of sound sources. In this paper, we propose a virtual microphone technique, which simulates binaural signals by equalizing ICTDs to ITDs, to localize sound images at azimuths of the sound sources with reference to the real microphones. Binaural signals simulated by the proposed method were examined objectively and subjectively by tests on two-sound-image localization. The tests revealed that the two sound images were localized at azimuths of the sound sources with reference to the real microphones.","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67081257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Takeru Misugi, Hideyoshi Miura, K. Hirata, T. Tachibana
{"title":"Design of Multiple Routing Configurations Considering Load Distribution for Network Slicing","authors":"Takeru Misugi, Hideyoshi Miura, K. Hirata, T. Tachibana","doi":"10.1561/116.00000148","DOIUrl":"https://doi.org/10.1561/116.00000148","url":null,"abstract":"","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67081632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this work, we focus on lightweight and accurate face alignment. For that purpose, we propose an algorithm design that promotes a most recently published face alignment method in terms of model size and computing cost while maintaining high accuracy of face alignment. Specifically, we construct a lightweight two-stage neural network. The first stage estimates boundary heatmaps on the facial region, which are then used to guide the facial landmark position prediction in the second stage. For the first stage, we compress an HourglassNet-based structure by reducing the numbers of feature channels and convolutional kernels and optimizing the structure of Hourglass block by ShuffleNet modules. For the second stage, we compress the subnet by utilizing DeLighT, a recently published lightweight version of Transformer. Experimental results on several standard facial landmark detection datasets show that the proposed algorithm achieves sharp advances in model compactness and computing efficiency while keeping a state-of-the-art level of accuracy in facial landmark detection.
{"title":"Lightweight Boundary-Aware Face Alignment with Compressed HourglassNet and Transformer","authors":"Wenhui Wang, Yingxin Li, Ziqiang Li, Jingliang Peng","doi":"10.1561/116.00000059","DOIUrl":"https://doi.org/10.1561/116.00000059","url":null,"abstract":"In this work, we focus on lightweight and accurate face alignment. For that purpose, we propose an algorithm design that promotes a most recently published face alignment method in terms of model size and computing cost while maintaining high accuracy of face alignment. Specifically, we construct a lightweight two-stage neural network. The first stage estimates boundary heatmaps on the facial region, which are then used to guide the facial landmark position prediction in the second stage. For the first stage, we compress an HourglassNet-based structure by reducing the numbers of feature channels and convolutional kernels and optimizing the structure of Hourglass block by ShuffleNet modules. For the second stage, we compress the subnet by utilizing DeLighT, a recently published lightweight version of Transformer. Experimental results on several standard facial landmark detection datasets show that the proposed algorithm achieves sharp advances in model compactness and computing efficiency while keeping a state-of-the-art level of accuracy in facial landmark detection.","PeriodicalId":44812,"journal":{"name":"APSIPA Transactions on Signal and Information Processing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135450441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：