The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
We propose a computer vision-based pipeline that enables altering the appearance of faces in videos. Assuming a surveillance scenario, we combine GMM-based background subtraction with an improved version of the GrabCut algorithm to find and segment pedestrians. Independently, we detect faces using a standard face detector. We apply the neural art algorithm, utilizing the responses of a deep neural...
In this paper, we propose a two-stage model for unconstrained face detection. The first stage is based on the normalized pixel difference (NPD) method, and the second stage uses the deformable part model (DPM) method. The NPD method applied to in the wild image datasets outputs the unbalanced ratio of false positive to false negative face detection when the main goal is to achieve minimal false negative...
In this paper, we present the Kernel Subclass Support Vector Data Description classifier. We focus on face recognition and human action recognition applications, where we argue that sub-classes are formed within the training class. We modify the standard SVDD optimization problem, so that it exploits subclass information in its optimization process. We extend the proposed method to work in feature...
The main drawback of speaker de-identification approaches using voice conversion techniques is the need for parallel corpora to train transformation functions between the source and target speakers. In this paper, a voice conversion approach that does not require training any parameters is proposed: it consists in manually defining frequency warping (FW) based transformations by using piecewise linear...
Assessment of the language environment of children in early childhood is a challenging task for both human and machine, and understanding the classroom environment of early learners is an essential step towards facilitating language acquisition and development. This paper explores an approach for intelligent language environment monitoring based on the duration of child-to-child and adult-to-child...
Many researchers have demonstrated the good performance of spoofing detection systems under clean training and testing conditions. However, it is well known that the performance of speaker and speech recognition systems significantly degrades in noisy conditions. Therefore, it is of great interest to investigate the effect of noise on the performance of spoofing detection systems. In this paper, we...
In this paper, an underwater acoustic sensor fault detection method is proposed that determines whether or not each sensor of multi-channel line array hydrophones malfunctions for passive sonar systems. To this end, the proposed method first measures a short-time root mean square (RMS) value of input signal for each channel. Then, it analyzes the RMS difference between the adjacent channels. In addition,...
Many approaches have been proposed to identify communities on complex networks. However the current algorithms are sensitive to the variation of input data and parameters. In this paper, we propose a new community detection approach called robust community detection on dynamic network (RCD). The robustness of our algorithm lies in two aspects. Firstly, by adopting the offset of sigmoid function, RCD...
In this paper, we present a cycled merging registration method based on Iterative Closest Point (ICP). We capture the point clouds by a static Kinect with the object rotating on a turntable. Different views of scan are combined by ICP and then a globally consistent human model is obtained. Our method simplifies the process of successively registration, which is usually used to solve multi-views registration...
Biometric system security requires cryptographic protection of sample data under certain circumstances. We introduce and assess low complexity selective encryption schemes applied to JPEG2000 compressed fingerprint data. From the results we are able to deduce design principles for such schemes which will guide to finally design recognition system aware encryption schemes with low encryption complexity...
In this paper we learn patterns of activity in open urban spaces and detect activity outliers that represent events of interest. We do so utilising background suppression to flag people as foreground blobs in videos from city surveillance cameras. Since the application domain is challenging, with far-field cameras viewing scenes that vary from completely empty to very crowded, and each person in the...
User authentication based on biometric information is an important part of computer security. A biometric system utilizes physiological or behavioural characteristics of a human being in order to confirm or to determine its identity. Handwriting as biometric modality can be based on the study of the writing result as well as on the observation of the writing process. While conventional sensors for...
Electroencephalographs (EEG) signals elicited by means of visual stimuli are highly time-dependent as they vary due to the subject's attention, state of mind, position of electrodes, etc., during acquisition. In this paper we exploit the use of techniques tailored to the analysis of signals varying across time. Specifically, dynamic time warping (DTW) is a technique to find an optimal alignment between...
This Paper proposes a novel method for the blind detection of image pre-processing techniques by means of statistical pattern recognition in image forensics. The technique is intended to detect sensor intrinsic pre-processing steps as well as manually applied filters. We have exemplary chosen 6 pre-processing filters with different parameter settings. The concept utilizes 29 image features which are...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.