2018 26th European Signal Processing Conference (EUSIPCO)最新文献_第9页

Colorblind-friendly Halftoning Colorblind-friendly半色调

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553352

S. K. F. Yu, Y. Chan, D. Lun, Chi Wang Jeffrey Chan, Kai Wang Kenneth Li

引用次数: 1

Active Learning for One-Class Classification Using Two One-Class Classifiers 基于两个单类分类器的单类分类主动学习

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8552958

Patrick Schlachter, Bin Yang

{"title":"Active Learning for One-Class Classification Using Two One-Class Classifiers","authors":"Patrick Schlachter, Bin Yang","doi":"10.23919/EUSIPCO.2018.8552958","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8552958","url":null,"abstract":"This paper introduces a novel, generic active learning method for one-class classification. Active learning methods play an important role to reduce the efforts of manual labeling in the field of machine learning. Although many active learning approaches have been proposed during the last years, most of them are restricted on binary or multi-class problems. One-class classifiers use samples from only one class, the so-called target class, during training and hence require special active learning strategies. The few strategies proposed for one-class classification either suffer from their limitation on specific one-class classifiers or their performance depends on particular assumptions about datasets like imbalance. Our proposed method bases on using two one-class classifiers, one for the desired target class and one for the so-called outlier class. It allows to invent new query strategies, to use binary query strategies and to define simple stopping criteria. Based on the new method, two query strategies are proposed. The provided experiments compare the proposed approach with known strategies on various datasets and show improved results in almost all situations.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123160310","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}

引用次数: 5

Emotion Estimation in Crowds: The Interplay of Motivations and Expectations in Individual Emotions 群体中的情绪估计:个人情绪中动机与期望的相互作用

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553370

Oscar J. Urizar, L. Marcenaro, C. Regazzoni, E. Barakova, G.W.M. Rauterberg

引用次数: 0

Skeleton-Based Action Recognition Based on Deep Learning and Grassmannian Pyramids 基于深度学习和格拉斯曼金字塔的骨骼动作识别

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553163

D. Konstantinidis, K. Dimitropoulos, P. Daras

引用次数: 9

A Fast Eigen-Based Signal Combining Algorithm by Using CORDIC 基于CORDIC的快速特征信号组合算法

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553406

Leiou Wang, Donghui Wang, C. Hao

引用次数: 1

Towards understanding the effects of practice on behavioural biometric recognition performance 了解练习对行为生物识别性能的影响

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553446

E. Haasnoot, J. S. Barnhoorrr, L. Spreeuwers, R. Veldhuis, W. Verwey

{"title":"Towards understanding the effects of practice on behavioural biometric recognition performance","authors":"E. Haasnoot, J. S. Barnhoorrr, L. Spreeuwers, R. Veldhuis, W. Verwey","doi":"10.23919/EUSIPCO.2018.8553446","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553446","url":null,"abstract":"Behavioural biometrics looks at discriminative features of a person's measurable behaviour, which is known to show high variance over long stretches of time. In psychology, a significant portion of this behavioural variance is explained by an individual improving their skill at performing behaviours, mostly through practice. Understanding what the effects of practice are on biometric recognition performance should allow us to account for much of this variance, as well as make individual behavioural biometric studies easier to compare [15]. We hypothesize that more accumulated practice will lead to both more stable and increased recognition performance. We argue that these are significant effects and show that practice in general is under-investigated. We introduce a novel method of analysis, the Start-to-Train Interval (STI)/Train-to-Test Interval (TTI) contour plot, which allows for systematic investigation of how recognition performance develops under increased practice. We applied this method to three data sets of a Discrete Sequence Production (DSP) task, a task that consists of repeatedly (500+ times) typing in a simple password, and found that more practice both significantly increases recognition performance and makes it more stable. These findings call for further investigation into the effects of practice on recognition performance for more standard behavioural biometric paradigms.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"156 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126920483","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}

引用次数: 3

Virtual camera modeling for multi-view simulation of surveillance scenes 多视点监控场景仿真的虚拟摄像机建模

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553409

N. Bisagno, N. Conci

引用次数: 3

Automated Tire Footprint Segmentation 自动轮胎足迹分割

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553041

R. Nava, D. Fehr, F. Petry, T. Tamisier

引用次数: 1

Playlist-Based Tag Propagation for Improving Music Auto-Tagging 基于播放列表的标签传播改进音乐自动标记

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553318

Yi-Hsun Lin, Chia-Hao Chung, Homer H. Chen

引用次数: 4

Modelling of Sound Events with Hidden Imbalances Based on Clustering and Separate Sub-Dictionary Learning 基于聚类和独立子字典学习的隐含不平衡声音事件建模

2018 26th European Signal Processing Conference (EUSIPCO) Pub Date : 2018-09-01 DOI: 10.23919/EUSIPCO.2018.8553387

Chaitanya Narisetty, Tatsuya Komatsu, Reishi Kondo

{"title":"Modelling of Sound Events with Hidden Imbalances Based on Clustering and Separate Sub-Dictionary Learning","authors":"Chaitanya Narisetty, Tatsuya Komatsu, Reishi Kondo","doi":"10.23919/EUSIPCO.2018.8553387","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553387","url":null,"abstract":"This paper proposes an effective modelling of sound event spectra with a hidden data-size-imbalance, for improved Acoustic Event Detection (AED). The proposed method models each event as an aggregated representation of a few latent factors, while conventional approaches try to find acoustic elements directly from the event spectra. In the method, all the latent factors across all events are assigned comparable importance and complexity to overcome the hidden imbalance of data-sizes in event spectra. To extract latent factors in each event, the proposed method employs clustering and performs non-negative matrix factorization to each latent factor, and learns its acoustic elements as a sub-dictionary. Separate sub-dictionary learning effectively models the acoustic elements with limited data-sizes and avoids over-fitting due to hidden imbalances in training data. For the task of polyphonic sound event detection from DCASE 2013 challenge, an AED based on the proposed modelling achieves a detection F-measure of 46.5%, a significant improvement of more than 19% as compared to the existing state-of-the-art methods.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"154 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121398008","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}

引用次数: 1