2020 8th International Workshop on Biometrics and Forensics (IWBF)最新文献_第3页

Bilateral Symmetry in Central Retinal Blood Vessels* 视网膜中央血管的双侧对称性*

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-04-01 DOI: 10.1109/IWBF49977.2020.9107969

S. Biswas, Johan Rohdin, M. Drahanský

引用次数: 1

Development of a score-to-likelihood ratio model for facial recognition using authentic criminalistic data 基于真实犯罪数据的面部识别得分似然比模型的开发

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-04-01 DOI: 10.1109/IWBF49977.2020.9107954

A. Mölder, Isabelle Enlund Åström, E. Leitet

引用次数: 5

Towards adept hand-crafted features for ocular biometrics 向熟练的手工制作特征的眼部生物识别

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-04-01 DOI: 10.1109/IWBF49977.2020.9107952

Ritesh Vyas

引用次数: 1

Doctoral Consortium Proposal: : Biometrics as forensic evidence: some reflections from the Italian Criminal proceeding’s point of view 博士联合会提案::生物特征作为法医证据:从意大利刑事诉讼程序角度的一些思考

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-04-01 DOI: 10.1109/IWBF49977.2020.9107948

Ernestina Sacchetto

{"title":"Doctoral Consortium Proposal: : Biometrics as forensic evidence: some reflections from the Italian Criminal proceeding’s point of view","authors":"Ernestina Sacchetto","doi":"10.1109/IWBF49977.2020.9107948","DOIUrl":"https://doi.org/10.1109/IWBF49977.2020.9107948","url":null,"abstract":"Biometric science is not free of errors, especially considering its typical statistical – probabilistic nature. The systematic use of biometric technologies in the Italian criminal proceeding could lead to some issues in terms of the reliability of the results from their application and the compatibility between the discipline in question, constitutional principles and typical procedural guarantees. However, because of the speed of the recent technology development, the criminal proceeding seems not to be able to do without the contribution offered by biometric science and the reason is that both the science and the trial, even with different approaches, have the different objective of the reconstruction of the causal connection. Biometric data have to be analyzed and evaluated in terms of scientific accuracy and it is necessary to understand their value case by case. The most crucial risks for forensic science in its recent digital dimension need to have legislative clarification aimed at standardising the discipline, also in terms of interpretability of technology, distinguishing one biometric data from another, implementing different applications since the specificity of each data element requires different legal solutions. Otherwise, the danger is that the procedural guarantees could be weakened (such as the principle of contradictory, the principle of reasonable duration of trial and the right of defence) in addition to some issues related to individual guarantees such as the right to privacy.","PeriodicalId":174654,"journal":{"name":"2020 8th International Workshop on Biometrics and Forensics (IWBF)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127278046","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: 0

Use of Scene Geometry Priors for Data Association in Egocentric Views 在自我中心视图中使用场景几何先验进行数据关联

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-04-01 DOI: 10.1109/IWBF49977.2020.9107955

Huiqin Chen, Emanuel Aldea, S. L. Hégarat-Mascle, V. Despiegel

引用次数: 1

Can GAN Generated Morphs Threaten Face Recognition Systems Equally as Landmark Based Morphs? - Vulnerability and Detection GAN生成的变形能像基于地标的变形一样威胁人脸识别系统吗?-漏洞及侦测

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-04-01 DOI: 10.1109/IWBF49977.2020.9107970

S. Venkatesh, Haoyu Zhang, Raghavendra Ramachandra, K. Raja, N. Damer, C. Busch

引用次数: 56

A Method for Curation of Web-Scraped Face Image Datasets 一种网络抓取人脸图像数据集的管理方法

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-04-01 DOI: 10.1109/IWBF49977.2020.9107950

Kai Zhang, Vítor Albiero, K. Bowyer

{"title":"A Method for Curation of Web-Scraped Face Image Datasets","authors":"Kai Zhang, Vítor Albiero, K. Bowyer","doi":"10.1109/IWBF49977.2020.9107950","DOIUrl":"https://doi.org/10.1109/IWBF49977.2020.9107950","url":null,"abstract":"Web-scraped, in-the-wild datasets have become the norm in face recognition research. The numbers of subjects and images acquired in web-scraped datasets are usually very large, with number of images on the millions scale. A variety of issues occur when collecting a dataset in-the-wild, including images with the wrong identity label, duplicate images, duplicate subjects and variation in quality. With the number of images being in the millions, a manual cleaning procedure is not feasible. But fully automated methods used to date result in a less-than-ideal level of clean dataset. We propose a semi-automated method, where the goal is to have a clean dataset for testing face recognition methods, with similar quality across men and women, to support comparison of accuracy across gender. Our approach removes near-duplicate images, merges duplicate subjects, corrects mislabeled images, and removes images outside a defined range of pose and quality. We conduct the curation on the Asian Face Dataset (AFD) and VGGFace2 test dataset. The experiments show that a state-of-the-art method achieves a much higher accuracy on the datasets after they are curated. Finally, we release our cleaned versions of both datasets to the research community.","PeriodicalId":174654,"journal":{"name":"2020 8th International Workshop on Biometrics and Forensics (IWBF)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131242236","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: 4

ComSupResNet: A Compact Super-Resolution Network for Low-Resolution Face Images ComSupResNet:用于低分辨率人脸图像的紧凑型超分辨率网络

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-04-01 DOI: 10.1109/IWBF49977.2020.9107946

Aashish Rai, Vishal M. Chudasama, Kishor P. Upla, K. Raja, Raghavendra Ramachandra, C. Busch

{"title":"ComSupResNet: A Compact Super-Resolution Network for Low-Resolution Face Images","authors":"Aashish Rai, Vishal M. Chudasama, Kishor P. Upla, K. Raja, Raghavendra Ramachandra, C. Busch","doi":"10.1109/IWBF49977.2020.9107946","DOIUrl":"https://doi.org/10.1109/IWBF49977.2020.9107946","url":null,"abstract":"Typically in face recognition based applications, a certain degree of resolution is required for effective feature extraction and comparison. Many practical systems such as surveillance applications violate the requirement by capturing Low-Resolution (LR) face images due to wider angle of imaging or longer stand-off to the camera. Such gap in the requirement versus the practice has lead numerous works to investigate approaches to super-resolve the face images that span from classical dictionary based methods to recent deep learning based approaches. In this work, we propose a compact and computationally efficient Convolutional Neural Network (CNN) to increase the spatial resolution of a LR face image to obtain a High-Resolution (HR) face image with an upscaling factor of up to ×8 which we refer as ComSupResNet. Contrary to other earlier works, the proposed architecture in the compact network comprises of a progressive residual propagating asym-metrical architecture with three modules: low-frequency and high-frequency feature extraction modules and reconstruction module. In addition to designing a new architecture, we also exercise care to reduce the number of parameters to approx. 1 M as compared to similar earlier work which has more than 30 M parameters. As a second aspect, we present the generalization aspect for the learned network in a cross-database setting by training the network on CelebA dataset while evaluating it both on CelebA and LFW datasets. Through empirical evaluations, we demonstrate the gain in high fidelity reconstruction in terms of structural similarity and Peak-Signal-to-Noise Ratio (PSNR) despite the compactness of the model.","PeriodicalId":174654,"journal":{"name":"2020 8th International Workshop on Biometrics and Forensics (IWBF)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121562215","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

ActGAN: Flexible and Efficient One-shot Face Reenactment ActGAN:灵活高效的一次性面部再现

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-03-30 DOI: 10.1109/IWBF49977.2020.9107944

Ivan Kosarevych, Marian Petruk, Markian Kostiv, Orest Kupyn, M. Maksymenko, Volodymyr Budzan

{"title":"ActGAN: Flexible and Efficient One-shot Face Reenactment","authors":"Ivan Kosarevych, Marian Petruk, Markian Kostiv, Orest Kupyn, M. Maksymenko, Volodymyr Budzan","doi":"10.1109/IWBF49977.2020.9107944","DOIUrl":"https://doi.org/10.1109/IWBF49977.2020.9107944","url":null,"abstract":"This paper introduces ActGAN – a novel end-to-end generative adversarial network (GAN) for one-shot face reenactment. Given two images, the goal is to transfer the facial expression of the source actor onto a target person in a photo-realistic fashion. While existing methods require target identity to be predefined, we address this problem by introducing a \"many-to-many\" approach, which allows arbitrary persons both for source and target without additional retraining. To this end, we employ the Feature Pyramid Network (FPN) as a core generator building block – the first application of FPN in face reenactment, producing finer results. We also introduce a solution to preserve a person’s identity between synthesized and target person by adopting the state-of-the-art approach in deep face recognition domain. The architecture readily supports reenactment in different scenarios: \"many-to-many\", \"one-to-one\", \"one-to-another\" in terms of expression accuracy, identity preservation, and overall image quality. We demonstrate that ActGAN achieves competitive performance against recent works concerning visual quality.","PeriodicalId":174654,"journal":{"name":"2020 8th International Workshop on Biometrics and Forensics (IWBF)","volume":"149 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133830871","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

Which Ear Regions Contribute to Identification and to Gender Classification? 哪些耳区有助于识别和性别分类?

2020 8th International Workshop on Biometrics and Forensics (IWBF) Pub Date : 2020-03-28 DOI: 10.1109/IWBF49977.2020.9107963

Di Meng, S. Mahmoodi, M. Nixon

引用次数: 4