Anil Kumar Sao

Face Verification Using Template Matching

Human faces are similar in structure with minor differences from person to person. These minor differences may average out while trying to synthesize the face image of a given person, or while building a model of face image in automatic face recognition. In this paper, we propose a template-matching approach for face verification, which neither synthesizes the face image nor builds a model of the face image. Template matching is performed using an edginess-based representation of the face image. The edginess-based representation of face images is computed using 1-D processing of images. An approach is proposed based on autoassociative neural network models to verify the identity of a person. The issues of pose and illumination in face verification are addressed.

On the use of phase of the Fourier transform for face recognition under variations in illumination

In this paper, we propose a representation of the face image based on the phase of the 2-D Fourier transform of the image to overcome the adverse effect of illumination. The phase of the Fourier transform preserves the locations of the edges of a given face image. The main problem in the use of the phase spectrum is the need for unwrapping of the phase. The problem of unwrapping is avoided by considering two functions of the phase spectrum rather than the phase directly. Each of these functions gives partial evidence of the given face image. The effect of noise is reduced by using the first few eigenvectors of the eigenanalysis on the two phase functions separately. Experimental results on combining the evidences from the two phase functions show that the proposed method provides an alternative representation of the face images for dealing with the issue of illumination in face recognition.

Significance of image representation for face verification

In this paper we discuss the significance of representation of images for face verification. We consider three different representations, namely, edge gradient, edge orientation and potential field derived from the edge gradient. These representations are examined in the context of face verification using a specific type of correlation filter, called the minimum average correlation energy (MACE) filter. The different representations are derived using one-dimensional (1-D) processing of image. The 1-D processing provides multiple partial evidences for a given face image, one evidence for each direction of the 1-D processing. Separate MACE filters are used for deriving each partial evidence. We propose a method to combine the partial evidences obtained for each representation using an auto-associative neural network (AANN) model, to arrive at a decision for face verification. Results show that the performance of the system using potential field representation is better than that using the edge gradient representation or the edge orientation representation. Also, the potential field representation derived from the edge gradient is observed to be less sensitive to variation in illumination compared to the gray level representation of images.

Edge preserving single image super resolution in sparse environment

Quality of an image is associated with edge of the image. It is important to preserve the edge of the image while deriving high resolution (HR) image from low resolution (LR) image, also known as superresolution (SR) problem. This paper proposes an edge preserving constraint, which preserve the edge information of image by minimizing the differences between edges of LR image and the edges of the reconstructed image (down-sampled version), in sparse coding based SR problem. Partial edge evidences, derived using 1-D processing of image, are used separately in the constraints. The experimental results show that proposed approach preserves the edges of image as well as outperforms objectively the existing SR approaches.

Greedy dictionary learning for kernel sparse representation based classifier

Proposed a novel kernel dictionary learning algorithm.Dictionary is updated in the coefficient domain instead of the signal domain.Proposed a hierarchical learning framework for efficient sparse representation.Proposed algorithm has much less computational complexity.Proposed approach performs well for various pattern classification tasks. We present a novel dictionary learning (DL) approach for sparse representation based classification in kernel feature space. These sparse representations are obtained using dictionaries, which are learned using training exemplars that are mapped into a high-dimensional feature space using the kernel trick. However, the complexity of such approaches using kernel trick is a function of the number of training exemplars. Hence, the complexity increases for large datasets, since more training exemplars are required to get good performance for most of the pattern classification tasks. To address this, we propose a hierarchical DL approach which requires the kernel matrix to update the dictionary atoms only once. Further, in contrast to the existing methods, the dictionary is learned in a linearly transformed/coefficient space involving sparse matrices, rather than the kernel space. Compared to the existing state-of-the-art methods, the proposed method has much less computational complexity, but performs similar for various pattern classification tasks.

A syllable-based framework for unit selection synthesis in 13 Indian languages

In this paper, we discuss a consortium effort on building text to speech (TTS) systems for 13 Indian languages. There are about 1652 Indian languages. A unified framework is therefore attempted required for building TTSes for Indian languages. As Indian languages are syllable-timed, a syllable-based framework is developed. As quality of speech synthesis is of paramount interest, unit-selection synthesizers are built. Building TTS systems for low-resource languages requires that the data be carefully collected an annotated as the database has to be built from the scratch. Various criteria have to addressed while building the database, namely, speaker selection, pronunciation variation, optimal text selection, handling of out of vocabulary words and so on. The various characteristics of the voice that affect speech synthesis quality are first analysed. Next the design of the corpus of each of the Indian languages is tabulated. The collected data is labeled at the syllable level using a semiautomatic labeling tool. Text to speech synthesizers are built for all the 13 languages, namely, Hindi, Tamil, Marathi, Bengali, Malayalam, Telugu, Kannada, Gujarati, Rajasthani, Assamese, Manipuri, Odia and Bodo using the same common framework. The TTS systems are evaluated using degradation Mean Opinion Score (DMOS) and Word Error Rate (WER). An average DMOS score of ≈3.0 and an average WER of about 20 % is observed across all the languages.

Voiced/nonvoiced detection in compressively sensed speech signals

Abstract We leverage the recent algorithmic advances in compressive sensing (CS), and propose a novel unsupervised voiced/nonvoiced (V/NV) detection method for compressively sensed speech signals. It attempts to exploit the fact that there is significant glottal activity during production of voiced speech while the same is not true for nonvoiced speech. This characteristic of the speech production mechanism is captured in the sparse feature vector derived using CS framework. Further, we propose an information theoretic metric, for V/NV classification, exploiting the sparsity of the extracted feature using a signal adaptive dictionary motivated by speech production mechanism. The final classification is done using an adaptive threshold selection scheme, which uses the temporal information of speech signals. While existing methods of feature extraction use speech samples directly, proposed method performs V/NV detection in compressively sensed speech signals (requiring very less memory), where existing time or frequency domain detection methods are not directly applicable. Hence, this method can be effective for various speech applications. Performance of the proposed method is studied on CMU-ARCTIC database, for eight types of additive noises, taken from the NOISEX database, at different signal-to-noise ratios (SNRs). The proposed method performs similar or better compared to the existing methods, especially at lower SNRs and this provide compelling evidence of the effectiveness of sparse feature vector for V/NV detection.

Noise adaptive super-resolution from single image via non-local mean and sparse representation

Abstract Super-resolution from a single image is a challenging task, more so, in presence of noise with unknown strength. We propose a robust super-resolution algorithm which adapts itself based on the noise-level in the image. We observe that dependency among the gradient values of relatively smoother patches diminishes with increasing strength of noise. Such a dependency is quantified using the ratio of first two singular values computed from local image gradients. The ratio is inversely proportional to the strength of noise. The number of patches with smaller ratio increases with increasing strength of noise. This behavior is used to formulate some parameters that are used in two ways in a sparse-representation based super-resolution approach: i) in computing an adaptive threshold, used in estimating the sparse coefficient vector via the iterative thresholding algorithm, ii) in choosing between the components representing image details and non-local means of similar patches. Furthermore, our approach constructs dictionaries by coarse-to-fine processing of the input image, and hence does not require any external training images. Additionally, an edge preserving constraint helps in better edge retention. As compared to state-of-the-art approaches, our method demonstrates better efficacy for optical and range images under different types and strengths of noise.

Hierarchical example-based range-image super-resolution with edge-preservation

We propose an example-based approach for enhancing resolution of range-images. Unlike most existing methods on range-image superresolution (SR), we do not employ a colour image counterpart for the range-image. Moreover, we use only a small set of range-images to construct a dictionary of exemplars. Considering the importance of edges in range-image SR, our formulation involves an edge-based constraint to better weight appropriate patches from the dictionary in a sparse-representation framework. Moreover, realizing the need for large up-sampling factors in case of range-images, we follow a hierarchical strategy for estimating the high-resolution range-images. We demonstrate that our strategy yields considerable improvements over the state-of-the-art approaches for range-image SR.

Determination of pose angle of face using dynamic space warping

In this paper we consider the problem of estimating the angle of pose of a face image with respect to frontal face image. Although the pose of a persons face depend on the 3D nature of the images, we show that it is possible to derive the pose of a face even from a 2D face image. This is primarily because there are several features of the image which are constrained because of it is a face. Also the objective is only to extract the single parameter pose angle, and not to reconstruct the face image. We use the features extracted along vertical scan line of an image to derive the pose angle. Using these vertical face features the image at any given pose is matched with face features from face image with front view, which corresponds to the zero pose. The matching is accomplished using dynamic space warping which is similar to dynamic time warping (DTW) used in matching the speech spectrum of isolated word recognition. The warping path obtained using DSW can be calibrated so that from the path one can derive the pose angle. In order to obtain good/accurate estimation of the pose angle, it is useful to have the reference image at several known angles.