Gayathri Mahalingam

Sample sequencing of vascular plants demonstrates widespread conservation and divergence of microRNAs

Small RNAs are pivotal regulators of gene expression that guide transcriptional and post-transcriptional silencing mechanisms in eukaryotes, including plants. Here we report a comprehensive atlas of sRNA and miRNA from 3 species of algae and 31 representative species across vascular plants, including non-model plants. We sequence and quantify sRNAs from 99 different tissues or treatments across species, resulting in a data set of over 132 million distinct sequences. Using miRBase mature sequences as a reference, we identify the miRNA sequences present in these libraries. We apply diverse profiling methods to examine critical sRNA and miRNA features, such as size distribution, tissue-specific regulation and sequence conservation between species, as well as to predict putative new miRNA sequences. We also develop database resources, computational analysis tools and a dedicated website, http://smallrna.udel.edu/. This study provides new insights on plant sRNAs and miRNAs, and a foundation for future studies.

Age invariant face recognition using graph matching

In this paper, we present a graph based face representation for efficient age invariant face recognition. The graph contains information on the appearance and geometry of facial feature points. An age model is learned for each individual and a graph space is built using the set of feature descriptors extracted from each face image. A two-stage method for matching is developed, where the first stage involves a Maximum a Posteriori solution based on PCA factorization to efficiently prune the search space and select very few candidate model sets. A simple deterministic algorithm which exploits the topology of the graphs is used for matching in the second stage. The experimental results on the FGnet database show that the proposed method is robust to age variations and provides better performance than existing techniques.

Face verification with aging using AdaBoost and local binary patterns

In this paper, we study the face verification task across age by constructing a simple but powerful representation of the face which uses Local Binary Pattern (LBP) histograms. The spatial information is incorporated by constructing a hierarchical representation of the face image and computing the LBP histogram at each level. A set of most discriminative LBP features of the face are extracted using the AdaBoost learning algorithm. A strong classifier is built using a set of weak classifiers extracted and is used for classification purposes. Several experiments on the FGnet and the MORPH database were performed and the results indicate a significant improvement in the performance when compared with other discriminative approaches. Performance improvement is achieved with smaller age gaps between image pairs and it stabilizes as the age gap increases. Also, the facial hair, glasses, etc. provide discriminative cues to the system in face verification.

Computational Methods for Comparative Analysis of Plant Small RNAs

Small RNAs play an important role in plant development, stress responses, and epigenetic regulation, primarily through their role in transcriptional and post-transcriptional silencing of specific target genes and loci. Most if not all plants utilize these small RNA signaling networks. We have developed a deep-sequencing based dataset of plant small RNAs, based on the hypothesis that comparisons among the complex pool of small RNAs from diverse plants will identify novel types of conserved, regulated, or species-specific molecules. A database containing upward of hundreds of millions of plant small RNA sequences is being created for comparative analyses. This small RNA database will allow the experimental characterization of the majority of the biologically important small RNAs for a range of plant species. This database can be accessed from our website (http://smallrna.udel.edu/). A variety of web-based tools have been developed for analyses of these data. Here, we focus on these tools, and we describe how the users can implement these tools to analyze and interpret the small RNA data and how the users could use similar approaches for other sets of plant small RNAs from diverse species.

Face verification of age separated images under the influence of internal and external factors

In this paper we study the task of face verification of age-separated images with the presence of various internal and external factors. We propose a hierarchical local binary pattern (HLBP) feature descriptor for robust face representation across age. The effective representation by HLBP across minimal age, illumination, and expression variations combined with its hierarchical computation provides a discriminative representation of the face image. The proposed face descriptor is combined with an AdaBoost classification framework to model the face verification task as a two-class problem. Experimental results on the FG-NET and MORPH aging datasets indicate that the performance of the proposed framework is robust with respect to images of both adults and children. A detailed empirical analysis on the effects of internal (age gap, gender, and ethnicity) and external (pose, expressions, facial hair, and glasses) factors in the face verification performance is also studied. The results indicate that the verification accuracy reduces as the age gap between the image pair increases. A quantitative comparison on the effects of gender on verification performance by both humans and the proposed machine learning approach is provided. The analysis indicate that the cues aid humans in verifying image pairs with large age gaps, while it aids machines for all age gaps. However, the cues mislead humans in the case of images of children and extra-personal pairs with large age gaps. Our analyses indicate that the pose and expression variations affect the performance, despite training with such variations, while facial hair and glasses act as discriminative cues. A study on the effects of ethnicity indicate that non-linear algorithms have insignificant effect in performance with the use of both generalized and individual ethnicity models when compared with linear algorithms.

Can discriminative cues aid face recognition across age

In this paper, we study the effects of discriminative cues (gender and age) in an image based face recognition system across age. We propose a pipeline framework to prune the search space based on gender and age of the test image to aid the process of recognition. A feature based approach which uses the Gabor phase and magnitude images of the face image is used in face image representation. The Gabor phases and magnitudes are encoded through Local Gabor Binary Pattern (LGBP) histograms. The gender and age group information of the face image is extracted using a random forest classifier based on a confidence measure and are used to prune the search space for efficient recognition. The effects of these discriminative cues are studied using a simple face recognition system in which the recognition is performed by histogram intersection technique which computes a similarity score between the LGBP histograms of two face images. The experiments on the FG-NET dataset and our private dataset show that the discriminative cues can indeed improve the performance of a face recognition system in terms of accuracy, lower time requirements, and graceful degradation.

Video based face recognition using graph matching

In this paper, we propose a novel graph based approach for still-to-video based face recognition, in which the temporal and spatial information of the face from each frame of the video is utilized. The spatial information is incorporated using a graph based face representation. The graphs contain information on the appearance and geometry of facial feature points and are labeled using the feature descriptors of the feature points. The temporal information is captured using an adaptive probabilistic appearance model. The recognition is performed in two stages where in the first stage a Maximum a Posteriori solution based on PCA is computed to prune the search space and select fewer candidates. A simple deterministic algorithm which exploits the topology of the graph is used for matching in the second stage. The experimental results on the UTD database and our dataset show that the adaptive matching and the graph based representation provides robust performance in recognition.

Face recognition in videos: a graph based modified kernel discriminant analysis

Grassmannian manifolds have been an effective way to represent image sets (video) which are mapped as data points on the manifold. Recognition can then be performed by applying the Discriminant Analysis (DA) on such manifolds. However, the local structure of the data points are not exploited in the DA. This paper proposes a modified Kernel Discriminant Analysis (KDA) approach on Grassmannian manifolds that utilizes the local structure of the data points on the manifold. The KDA exploits the local structure using between-class and within-class adjacency graphs that represent the between-class and within-class similarities, respectively. The maximum correlation from within-class and minimum correlation from between-class is utilized to define the connectivity between points in the graph thus exploiting the geometrical structure of the data. The discriminability is further improved by effective feature representation using LBP which can discriminate data across illumination, pose, and minor expressions. Effective recognition is performed by using only the cluster representatives extracted by clustering the frames of a video sequence. Experiments on several video datasets (Honda, MoBo, ChokePoint, NRC-IIT, and MOBIO) show that the proposed approach obtains better recognition rates, in comparison with the state-of-the-art approaches.

Face recognition in videos using adaptive graph appearance models

In this paper, we present a novel graph, sub-graph and supergraph based face representation which captures the facial shape changes and deformations caused due to pose changes and use it in the construction of an adaptive appearance model. This work is an extension of our previous work proposed in [1]. A sub-graph and super-graph is extracted for each pair of training graphs of an individual and added to the graph model set and used in the construction of appearance model. The spatial properties of the feature points are effectively captured using the graph model set. The adaptive graph appearance model constructed using the graph model set captures the temporal characteristics of the video frames by adapting the model with the results of recognition from each frame during the testing stage. The graph model set and the adaptive appearance model are used in the two stage matching process, and are updated with the sub-graphs and super-graphs constructed using the graph of the previous frame and the training graphs of an individual. The results indicate that the performance of the system is improved by using subgraphs and super-graphs in the appearance model.