Kishor S. Kinage

Bio: Kishor S. Kinage is an academic researcher from College of Engineering, Pune. The author has contributed to research in topics: Feature extraction & Deep belief network. The author has an hindex of 2, co-authored 8 publications receiving 13 citations.

Comprehensive review on facial based human age estimation

Abstract: Recently facial based age estimation has become increasingly important because of many potential real time applications. Age estimation is predicting someone's age by analyzing his/her biometric trait such as bone density, dental structure or face. Amongst these face is important trait so facial based age estimation has become more popular due to its vast real time applications. Age estimation is defined as to label the face image automatically with the exact age or age group. Estimating age from images has been one of the most challenging problems within the field of facial analysis due to uncontrollable nature of the aging process, high variance of observations within the same age range, lighting, facial expressions, pose, occlusion, blur, camouflage due to beards, moustache, glasses, makeup and the difficulty to gather complete and sufficient training data. This paper presents analysis of earlier techniques proposed by researchers for facial based age estimation. Different feature extraction and estimator learning methods used in this domain are also discussed.

Iris Tracking Using a Single Web-Cam without IR Illumination

Abstract: Iris Tracking is the process of determining the point of gaze or the motion of an eye. In today's era, the combination of Iris Tracking and gaze estimation shows a person's interest. In this paper, we have focused on a single-camera-based gaze estimation algorithm. The paper also describes the implementation of both iris and movement of cursor according to iris position which can be used and detected for gaze estimation in order to improve accuracy without using IR Illumination or any sensor camera. The system works at different distraction conditions, as the normalized error rate is minimal.

Functional Verification of Ten Gigabytes Media Independent Interface (XGMII) Using Universal Verification Methodology

Abstract: To provide a simple, inexpensive, and easy to implement interconnection between the MAC sublayer and PHY, XGMII can be implement. Since XGMII is an optional interface, it is used in the 10-Gigabit Ethernet standard as the basis for the specification. Functional Verification of XGMII (Ten Giga-bytes Media Independent Interface) using advanced Verification Methodology which is UVM (Universal Verification Methodology), is presented in this paper. The main aspect of MAC is to sending Ethernet frames to PHY through XGMII and to receive that frames from PHY to MAC sublayer through the same interface. Coverage driven verification is effectively achieved by UVM with the use of factory and configuration mechanism, coverage metrics and self-checking which reduces the time spent on verifying the design. By using UVM methodology a reusable testbench is developed, which has been used to run different test scenarios on same TB environment. Regression testing of the design is carried out for achieving better coverage goal.

CDBN: Crow Deep Belief Network Based on Scattering and AAM Features for Age Estimation

Abstract: Automatic age estimation from the face images is a growing research interest nowadays. Various literature works have contributed towards the age detection scheme, besides only a few have resulted in providing good performance. This is due to the influence of the external factors, such as environment, lifestyle, and various expressions present in the face image. This paper proposes a deep belief network with the crow optimization algorithm for the age detection purpose. The proposed Crow Deep Belief Network (CDBN) finds the age of the person in the image through the initial training with the face features. The features for the training of the proposed CDBN are provided by the scattering transform and the Active Appearance Model (AAM). The training of the CDBN with the features provides the optimal weights used for the age detection. The experimentation of the proposed CDBN is done by four standard databases, namely the IMDB database, the Adience database, the AFAD database, and the FG-NET database based on the metrics, such as Mean Absolute Error (MAE), Accuracy of error of one age category (AEO) and Accuracy of an Exact Match (AEM). Among them, the proposed model has the minimum MAE with a value of 2.186 for FG-NET database, and maximum AEO and AEM with the values of 0.972, and 0.971, respectively for IMDB database.

Facial Based Human Age Estimation Using Deep Belief Network

Abstract: Facial based human age estimation has attracted lot of attention nowadays. Age estimation has become quite challenging task due to variation in lighting conditions, poses, and facial expression. Despite so much research in facial based human age estimation still there is room to improve performance. To improve accuracy we present age estimation using deep belief network. Deep belief network have shown superior performance as compared to other classification models. Success of deep belief network lies in contrastive divergence algorithm. Facial images passes though viola johns facial detection algorithm, once face is detected facial featured are extracted using active appearance and scattering transform feature method. These feature extraction model not only extracts geometric features but also extracts texture features. Subsequently deep belief network classification model is built on partitioned training images and evaluated on testing images. We performed experimentation on training images. Dataset and results are obtained by varying training percentages. Compared to other age estimation models we achieved low mean absolute error of 4.95 for 70% training images dataset. This study shows that due to inclusion of deep belief network performance is excelled.

Automatic facial age estimation

Abstract: The reliability of automatically estimating human ages, by processing input facial images, has generally been found to be poor. On other hand, various real world applications, often relating to safety and security, depend on an accurate estimate of a person’s age. In such situations, Face Image based Automatic Age Estimation (FI-AAE) systems which are more reliable and may ideally surpass human ability, are of importance as and represent a critical pre-requisite technology. Unfortunately, in terms of estimation accuracy and thus performance, contemporary FI-AAE systems are impeded by challenges which exist in both of the two major FI-AAE processing phases i.e. i) Age based feature extraction and representation and ii) Age group classification. Challenges in the former phase arise because facial shape and texture change independently and the magnitude of these changes vary during the different stages of a person’s life. Additionally, contemporary schemes struggle to exploit age group specific characteristics of these features, which in turn has a detrimental effect on overall system performance. Furthermore misclassification errors which occur in the second processing phase and are caused by the smooth inter-class variations often observed between adjacent age groups, pose another major challenge and are responsible for low overall FI-AAE performance. In this thesis a novel Multi-Level Age Estimation (ML-AE) framework is proposed that addresses the aforementioned challenges and improves upon state-of-the-art FI-AAE system performance. The proposed ML-AE is a hierarchical classification scheme that maximizes and then exploits inter-class variation among different age groups at each level of the hierarchy. Furthermore, the proposed scheme exploits age based discriminating information taken from two different cues (i.e. facial shape and texture) at the decision level which improves age estimation results. During the process of achieving our main objective of age estimation, this research work also contributes to two associated image processing/analysis areas: i) Face image modeling and synthesis; a process of representing face image data with a low dimensionality set of parameters. This is considered as precursor to every face image based age estimation system and has been studied in this thesis within the context of image face recognition ii) measuring face image data variability that can help in representing/ranking different face image datasets according to their classification difficulty level. Thus a variability measure is proposed that can also be used to predict the classification performance of a given face recognition system operating upon a particular input face dataset. Experimental results based on well-known face image datasets revealed the superior performance of our proposed face analysis, synthesis and face image based age classification methodologies, as compared to that obtained from conventional schemes.

Volterra modeling of the human smooth pursuit system in health and disease

Abstract: This thesis treats the identification of Volterra models of the human smooth pursuit system from eye-tracking data. Smooth pursuit movements are gaze movements used in tracking of moving targets an ...

Eye gaze location detection based on iris tracking with web camera

Abstract: In recent years, there has been an increased interest in human computer interaction systems where web cameras are used as input devices. In this study, a system developed to distinguish eye gaze locations on the screen by referring to the position of the iris part of the eye is introduced. It is aimed to distinguish more eye gaze location compared to the previous works by web cam. K-nearest neighbors classifier was used to detect eye gaze location by the feature vector of the iris center coordinates. As a result of the first experiments with 10 subjects, the 17 different eye gaze locations on the screen have classified with an average of 97.64% accuracy. It has been observed that only the two adjacent points near the center of the screen in the vertical direction are detected incorrectly. It is expected to increase the classification performance ratio by not using these two incorrectly detected points in future studies.

A Comprehensive Investigation of Universal Verification Methodology (UVM) Standard for Design Verification

Abstract: Universal Verification Methodology (UVM) is getting attention of researchers and functional verification community due to its advance flexibility, reusability and reliability features for design verification of multifaceted embedded systems. This is the reason that different tool vendors like Mentor Graphics support UVM-based simulation for design verification. Similarly, researchers frequently explore / utilize UVM to enhance the verification capabilities for embedded systems. In this context, there is a strong need to summarize the latest advancements, tools and techniques for UVM standard. Therefore, this article performs a Systematic Literature Review (SLR) to identify 27 studies (i.e. 2017-2019) pertaining to UVM standard. Subsequently, 21 UVM-based frameworks and 9 tools are identified. Moreover, key benefits of UVM standard are investigated. Finally, a comparative analysis of UVM with OVM (Open Verification Methodology) is performed. It is concluded that UVM provides advanced phasing mechanism, reporting, callbacks, objections, sequence libraries and control over simulation as compared to OVM. Researchers and practitioners of domain can highly benefit from findings of this article.

Deep Convolutional Neural Networks for Forensic Age Estimation: A Review

Abstract: Forensic age estimation is usually requested by courts, but applications can go beyond the legal requirement to enforce policies or offer age-sensitive services. Various biological features such as the face, bones, skeletal and dental structures can be utilised to estimate age. This article will cover how modern technology has developed to provide new methods and algorithms to digitalise this process for the medical community and beyond. The scientific study of Machine Learning (ML) have introduced statistical models without relying on explicit instructions, instead, these models rely on patterns and inference. Furthermore, the large-scale availability of relevant data (medical images) and computational power facilitated by the availability of powerful Graphics Processing Units (GPUs) and Cloud Computing services have accelerated this transformation in age estimation. Magnetic Resonant Imaging (MRI) and X-ray are examples of imaging techniques used to document bones and dental structures with attention to detail making them suitable for age estimation. We discuss how Convolutional Neural Network (CNN) can be used for this purpose and the advantage of using deep CNNs over traditional methods. The article also aims to evaluate various databases and algorithms used for age estimation using facial images and dental images.