Aditi Roy

Gait recognition using Pose Kinematics and Pose Energy Image

Many of the existing gait recognition approaches represent a gait cycle using a single 2D image called Gait Energy Image (GEI) or its variants. Since these methods suffer from lack of dynamic information, we model a gait cycle using a chain of key poses and extract a novel feature called Pose Energy Image (PEI). PEI is the average image of all the silhouettes in a key pose state of a gait cycle. By increasing the resolution of gait representation, more detailed dynamic information can be captured. However, processing speed and space requirement are higher for PEI than the conventional GEI methods. To overcome this shortcoming, another novel feature named as Pose Kinematics is introduced, which represents the percentage of time spent in each key pose state over a gait cycle. Although the Pose Kinematics based method is fast, its accuracy is not very high. A hierarchical method for combining these two features is, therefore, proposed. At first, Pose Kinematics is applied to select a set of most probable classes. Then, PEI is used on these selected classes to get the final classification. Experimental results on CMUs Mobo and USFs HumanID data set show that the proposed approach outperforms existing approaches.

Pose Depth Volume extraction from RGB-D streams for frontal gait recognition

Highlights? We combine depth and RGB information from Kinect for frontal gait recognition. ? Key poses are extracted using depth frames registered in RGB frame coordinate system. ? A new feature named Pose Depth Volume is proposed. ? Comparative study with existing gait features has been done. We explore the applicability of Kinect RGB-D streams in recognizing gait patterns of individuals. Gait energy volume (GEV) is a recently proposed feature that performs gait recognition in frontal view using only depth image frames from Kinect. Since depth frames from Kinect are inherently noisy, corresponding silhouette shapes are inaccurate, often merging with the background. We register the depth and RGB frames from Kinect to obtain smooth silhouette shape along with depth information. A partial volume reconstruction of the frontal surface of each silhouette is done and a novel feature termed as Pose Depth Volume (PDV) is derived from this volumetric model. Recognition performance of the proposed approach has been tested on a data set captured using Microsoft Kinect in an indoor environment. Experimental results clearly demonstrate the effectiveness of the approach in comparison with other existing methods.

A hierarchical method combining gait and phase of motion with spatiotemporal model for person re-identification

Re-identification refers to the problem of establishing correspondence among various observations of the same subject viewed at different time instances in different camera positions. We propose a hierarchical approach for re-identifying a subject by combining gait with phase of motion and a spatiotemporal model. The fundamental nature of the gait biometric of being amenable to capturing from a distance even at low resolution without active co-operation of subjects, has motivated us to use it for re-identification. We use two features related to a subjects motion dynamics, one is his exit/entry phase of motion and the other is his gait signature. An additional third feature is obtained from the spatiotemporal model of the camera network which is learnt during the training phase in the form of a multivariate probability density of space-time variables (entry/exit location, exit velocity, and inter-camera travel time) using kernel density estimation. Once all these three features have been computed, correspondences are established by dynamic programing based maximum likelihood (ML) estimation. The performance of our method has been evaluated on a real data set featuring a two-camera and a three-camera network in a hallway monitoring situation. The proposed approach shows promising results on both the data sets.

An HMM-based behavior modeling approach for continuous mobile authentication

This paper studies continuous authentication for touch interface based mobile devices. A Hidden Markov Model (HMM) based behavioral template training approach is presented, which does not require training data from other subjects other than the owner of the mobile. The stroke patterns of a user are modeled using a continuous left-right HMM. The approach models the horizontal and vertical scrolling patterns of a user since these are the basic and mostly used interactions on a mobile device. The effectiveness of the proposed method is evaluated through extensive experiments using the Toucha-lytics database which comprises of touch data over time. The results show that the performance of the proposed approach is better than the state-of-the-art method.

State-Based Modeling and Object Extraction From Echocardiogram Video

In this paper, we propose a hierarchical state-based model for representing an echocardiogram video. It captures the semantics of video segments from dynamic characteristics of objects present in each segment. Our objective is to provide an effective method for segmenting an echo video into view, state, and substate levels. This is motivated by the need for building efficient indexing tools to support better content management. The modeling is done using four different views, namely, short axis, long axis, apical four chamber, and apical two chamber. For view classification, an artificial neural network is trained with the histogram of a region of interest of each video frame. Object states are detected with the help of synthetic M-mode images. In contrast to traditional single M-mode, we present a novel approach named sweep M-mode for state detection. We also introduce radial M-mode for substate identification from color flow Doppler 2-D imaging. The video model described here represents the semantics of video segments using first-order predicates. Suitable operators have been defined for querying the segments. We have carried out experiments on 20 echo videos and compared the results with manual annotation done by two experts. View classification accuracy is 97.19%. Misclassification error of the state detection stage is less than 13%, which is within acceptable range since only frames at the state boundaries are found to be misclassified.

Design and Analysis of Shoulder Surfing Resistant PIN Based Authentication Mechanisms on Google Glass

This paper explores options to the built-in authentication mechanism of the Google Glass which is vulnerable to shoulder surfing attacks. Two simple PIN-based authentication techniques are presented, both of which provide protection against shoulder surfing. The techniques employ two interfaces for entering the PIN, namely, voice (Voice-based PIN) and touchpad (Touch-based PIN). To enter the same PIN, user has the freedom to choose either technique and thereby interface, as per the environment in which authentication is being performed. A user study was conducted with 30 participants to compare the performance of the proposed methods with the built-in technique. The results show that the proposed mechanisms have a significantly better login success rate than the built-in technique. Interestingly, although the average authentication times of the proposed methods are higher than that of the built-in one, the users perceived them as being faster. The results also indicate that the proposed methods have better perceived security and usability than the built-in method. The study reveals that when it comes to authentication on augmented reality devices, there is a need for authentication mechanisms that complement each other as users tend to prefer a different interface in different contexts.

MasterPrint: Exploring the Vulnerability of Partial Fingerprint-Based Authentication Systems

This paper investigates the security of partial fingerprint-based authentication systems, especially when multiple fingerprints of a user are enrolled. A number of consumer electronic devices, such as smartphones, are beginning to incorporate fingerprint sensors for user authentication. The sensors embedded in these devices are generally small and the resulting images are, therefore, limited in size. To compensate for the limited size, these devices often acquire multiple partial impressions of a single finger during enrollment to ensure that at least one of them will successfully match with the image obtained from the user during authentication. Furthermore, in some cases, the user is allowed to enroll multiple fingers, and the impressions pertaining to multiple partial fingers are associated with the same identity (i.e., one user). A user is said to be successfully authenticated if the partial fingerprint obtained during authentication matches any one of the stored templates. This paper investigates the possibility of generating a “MasterPrint,” a synthetic or real partial fingerprint that serendipitously matches one or more of the stored templates for a significant number of users. Our preliminary results on an optical fingerprint data set and a capacitive fingerprint data set indicate that it is indeed possible to locate or generate partial fingerprints that can be used to impersonate a large number of users. In this regard, we expose a potential vulnerability of partial fingerprint-based authentication systems, especially when multiple impressions are enrolled per finger.

Occlusion detection and gait silhouette reconstruction from degraded scenes

Gait, which is defined as the style of walking of a person, has been recognized as a potential biometric feature for identifying human beings. The fundamental nature of gait biometric of being unconstrained and captured often without a subject’s knowledge or co-operation has motivated many researchers over the last one decade. However, all of the approaches found in the literature assume that there is little or no occlusion present at the time of capturing gait images, both during training and during testing and deployment. We look into this challenging problem of gait recognition in the presence of occlusion. A novel approach is proposed, which first detects the presence of occlusion and accordingly extracts clean and unclean gait cycles from the whole input sequence. In the second step, occluded silhouette frames are reconstructed using Balanced Gaussian Process Dynamical Model (BGPDM). We evaluated our approach on a new data set TUM-IITKGP featuring inter-object occlusion. Algorithms have also been tested on CMU’s Mobo data set by introducing synthetic occlusion of different degrees. The proposed approach shows promising result on both the data sets.

An HMM-based multi-sensor approach for continuous mobile authentication

With the increased popularity of smart phones, there is a greater need to have a robust authentication mechanism that handles various security threats and privacy leakages effectively. This paper studies continuous authentication for touch interface based mobile devices. A Hidden Markov Model (HMM) based behavioral template training approach is presented, which does not require training data from other subjects other than the owner of the mobile device and can get updated with new data over time. The gesture patterns of the user are modeled from multiple sensors - touch, accelerometer and gyroscope data using a continuous left-right HMM. The approach models the tap and stroke patterns of a user since these are the basic and most frequently used interactions on a mobile device. To evaluate the effectiveness of the proposed method a new data set has been created from 42 users who interacted with off-the-shelf applications on their smart phones. Results show that the performance of the proposed approach is promising and potentially better than other state-of-the-art approaches.

Summarization of Neonatal Video EEG for Seizure and Artifact Detection

Monitoring neonatal EEG signal is useful in identifying neonatal convulsions or seizures. For neonates, seizures can be electrographic, electro clinical, or both simultaneously. Electrographic seizure is identified via recorded EEG signal, while electro clinical seizures exhibit clinical manifestations. Sometimes neonates can exhibit silent seizures which may be clinically invisible but identifiable in recorded EEG, or vice versa. Thus, simultaneous monitoring of video and recorded EEG determines the correlation between the electrographic and electro clinical seizures. Furthermore, analyzing the movements of the neonates can identify movement artifacts easily, thus preventing false seizure detection. However, storage of high quality video recordings require large storage space. As neonates do not commonly exhibit movements, summarizing the video for storing only patient movements along with corresponding timestamps, can be useful. In this paper, a video summarization method is proposed for efficient browsing of video-EEG. Identification and analysis of the patterns of interest is possible via summarized information, thus reducing effective analysis time. In addition, quantitative demonstration of electrographic and electro clinical seizures is presented to analyze the utility of video-EEG.