M. J. Gómez-Silva, Jose M. Armingol, A. D. L. Escalera
{"title":"Triplet Permutation Method for Deep Learning of Single-Shot Person Re-Identification","authors":"M. J. Gómez-Silva, Jose M. Armingol, A. D. L. Escalera","doi":"10.1049/cp.2019.1168","DOIUrl":"https://doi.org/10.1049/cp.2019.1168","url":null,"abstract":"Solving Single-Shot Person Re-Identification (Re-Id) by training Deep Convolutional Neural Networks is a daunting challenge, due to the lack of training data, since only two images per person are available. This causes the overfitting of the models, leading to degenerated performance. This paper formulates the Triplet Permutation method to generate multiple training sets, from a certain re-id dataset. This is a novel strategy for feeding triplet networks, which reduces the overfitting of the Single-Shot Re-Id model. The improved performance has been demonstrated over one of the most challenging Re-Id datasets, PRID2011, proving the effectiveness of the method.","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114869184","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}
Nima Mohammadi Meshky, Sara Iodice, K. Mikolajczyk
{"title":"Domain Adversarial Training for Infrared-colour Person Re-Identification","authors":"Nima Mohammadi Meshky, Sara Iodice, K. Mikolajczyk","doi":"10.1049/cp.2019.1167","DOIUrl":"https://doi.org/10.1049/cp.2019.1167","url":null,"abstract":"Person re-identification (re-ID) is a very active area of research in computer vision, due to the role it plays in video surveillance. Currently, most methods only address the task of matching between colour images. However, in poorly-lit environments CCTV cameras switch to infrared imaging, hence developing a system which can correctly perform matching between infrared and colour images is a necessity. In this paper, we propose a part-feature extraction network to better focus on subtle, unique signatures on the person which are visible across both infrared and colour modalities. To train the model we propose a novel variant of the domain adversarial feature-learning framework. Through extensive experimentation, we show that our approach outperforms state-of-the-art methods.","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114989771","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}
M. L. Durova, A. Dimou, George C. Litos, P. Daras, J. Davis
{"title":"TooManyEyes: Super-recogniser directed identification of target individuals on CCTV","authors":"M. L. Durova, A. Dimou, George C. Litos, P. Daras, J. Davis","doi":"10.5281/ZENODO.1071985","DOIUrl":"https://doi.org/10.5281/ZENODO.1071985","url":null,"abstract":"For the current research, a ‘Spot the Face in a Crowd Test’ (SFCT) comprising six video clips depicting target-actors and multiple bystanders was loaded on TooManyEyes, a bespoke multi-media platform adapted here for the human-directed identification of individuals in CCTV footage. To test the utility of TooManyEyes, police ‘super-recognisers’ (SRs) who may possess exceptional face recognition ability, and police controls attempted to identify the target-actors from the SFCT. As expected, SRs correctly identified more target-actors; with higher confidence than controls. As such, the TooManyEyes system provides a useful platform for uploading tests for selecting police or security staff for CCTV review deployment","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"225 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132915925","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":"Designing a facial spoofing database for processed image attacks","authors":"Luma Omar, I. Ivrissimtzis","doi":"10.1049/ic.2016.0073","DOIUrl":"https://doi.org/10.1049/ic.2016.0073","url":null,"abstract":"Face recognition systems are used for user authentication in everyday applications such as logging into a laptop or smartphone without need to memorize a password. However, they are still vulnerable to spoofing attacks, as for example when an imposter gains access to a system by holding a printed photo of the rightful user in front of the camera. In this paper we are concerned with the design of face image databases for evaluating the performance of anti-spoofing algorithms against such attacks. We present a new database, supporting testing against an enhancement of the attack, where the imposter processes the stolen image before printing it. By testing a standard antispoofing algorithm on the new database we show a significant decrease in its performance and, as a simple remedy to this problem, we propose the inclusion of processed imposter images into the training set.","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132937090","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":"Detecting acceleration for gait and crime scene analysis","authors":"Y. Sun, Jonathon S. Hare, M. Nixon","doi":"10.1049/IC.2016.0071","DOIUrl":"https://doi.org/10.1049/IC.2016.0071","url":null,"abstract":"Identifying criminals from CCTV footage is often a difficult task for crime investigations. The quality of CCTV is often low and criminals can cover their face and wear gloves (to withhold fingerprints) when committing a crime. Gait is the optimal choice in this circumstance since people can be recognised by their walking style, even at a distance with low resolution imagery. The location of the frame when the heel strikes the floor is essential for some gait analyses. We propose a new method to detect heel strikes: by radial acceleration which can also generalise to crime analysis. The frame and position of the heel strikes can be estimated by the quantity and the circle centres of radial acceleration, derived from the optical flow (using DeepFlow). Experimental results show high detection rate on two different gait databases and good robustness under different kinds of noise. We analysedetection of heel strikes to show robustness then we analyse crime scenes to show generalisation capability since violent crime often involves much acceleration. As such, we provide a new basis to a baseline technique in crime scene analysis.","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116365114","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}
Nicolas Jaccard, T. W. Rogers, E. Morton, Lewis D. Griffin
{"title":"Automated detection of smuggled high-risk security threats using Deep Learning","authors":"Nicolas Jaccard, T. W. Rogers, E. Morton, Lewis D. Griffin","doi":"10.1049/IC.2016.0079","DOIUrl":"https://doi.org/10.1049/IC.2016.0079","url":null,"abstract":"The security infrastructure is ill-equipped to detect and deter the smuggling of non-explosive devices that enable terror attacks such as those recently perpetrated in western Europe. The detection of so-called \"small metallic threats\" (SMTs) in cargo containers currently relies on statistical risk analysis, intelligence reports, and visual inspection of X-ray images by security officers. The latter is very slow and unreliable due to the difficulty of the task: objects potentially spanning less than 50 pixels have to be detected in images containing more than 2 million pixels against very complex and cluttered backgrounds. In this contribution, we demonstrate for the first time the use of Convolutional Neural Networks (CNNs), a type of Deep Learning, to automate the detection of SMTs in fullsize X-ray images of cargo containers. Novel approaches for dataset augmentation allowed to train CNNs from-scratch despite the scarcity of data available. We report fewer than 6% false alarms when detecting 90% SMTs synthetically concealed in stream-of-commerce images, which corresponds to an improvement of over an order of magnitude over conventional approaches such as Bag-of-Words (BoWs). The proposed scheme offers potentially super-human performance for a fraction of the time it would take for a security officers to carry out visual inspection (processing time is approximately 3.5s per container image).","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134165471","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":"Multisensor concealed weapon detection using the image fusion approach","authors":"T. Xu, Q. Wu","doi":"10.1049/IC.2015.0108","DOIUrl":"https://doi.org/10.1049/IC.2015.0108","url":null,"abstract":"In this paper, an efficient concealed weapon detection (CWD) algorithm based on image fusion is presented. First, the images obtained using different sensors are decomposed into low and high frequency bands with the double-density dual-tree complex wavelet transform (DDDTCWT). Then two novel decision methods are introduced referring to the characteristics of the frequency bands, which significantly improves the image fusion performance for CWD application. The fusion of low frequency bands coefficients is determined by the local contrast, while the high frequency band fusion rule is developed by considering both the texture feature of the human visual system (HVS) and the local energy basis. Finally, the fused image is obtained through the inverse DDDTCWT. Experiments and comparisons demonstrate the robustness and efficiency of the proposed approach and indicate that the fusion rules can be applied to different multiscale transforms. Also, our work shows that the fusion result using the proposed fusion rules on DDDTCWT is superior to other combinations as well as previously proposed approaches.","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133253113","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":"Automatic speech recognition enhancement using adaptive GMM estimation","authors":"Kemouche Abdennour, N. Aouf","doi":"10.1049/IC.2015.0117","DOIUrl":"https://doi.org/10.1049/IC.2015.0117","url":null,"abstract":"In this paper, we present an automatic speech recognition system based on an adaptive Gaussian mixture technique dealing with audio signal modality. To perform robust density estimation after speech feature extraction stage, an adaptive mixture estimation method is used based on optimal minimization of the integral square distance between the true density that represents the speech features and the approximated mixture. This estimation is relatively difficult because of the complex representation of the density and the issues with Expectation-Maximization (EM) algorithm classically used for these approximations. The technique we are proposing in this work not only shows its performance through the experimental results of this paper but also provides in the future a natural and efficient way of including bimodality (audio and video) into our robust automatic speech recognition program of study.","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126600189","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":"Retrieval of distinctive regions of interest from video surveillance footage: a real use case study","authors":"C. Mitrea, T. Piatrik, B. Ionescu, M. Neville","doi":"10.1049/IC.2015.0110","DOIUrl":"https://doi.org/10.1049/IC.2015.0110","url":null,"abstract":"The article addresses the issue of retrieving distinctive regions of interest or patterns (DROP) in video surveillance datasets. DROP may include logos, tattoos, color regions or any other distinctive features that appear recorded on video. These data come in particular with specific difficulties such as low image quality, multiple image perspectives, variable lighting conditions and lack of enough training samples. This task is a real need functionality as the challenges are derived from practice of police forces. We present our preliminary results on tackling such scenario from Scotland Yard, dealing with the constraints of a real world use case. The proposed method is based on two approaches: employment of a dense SIFT-based descriptor (Pyramidal Histogram of Visual Words), and use of image segmentation (Mean-Shift) with feature extraction on each segment computed. Tested on real data we achieve very promising results that we believe will contribute further to the ground development of advanced methods to be applied and tested in real forensics investigations.","PeriodicalId":215265,"journal":{"name":"International Conferences on Imaging for Crime Detection and Prevention","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129329415","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}