Qualification: 
Ph.D, M.Tech, B-Tech
sandeepp@am.amrita.edu

Dr. Sandeep P. currently serves as an Assistant Professor in the Department of Electronics and Communication Engineering, School of Engineering, Amrita Vishwa Vidyapeetham, Amritapuri campus.

He received his B.Tech. in Electronics and Communication Engineering from Government College of Engineering, Kannur, in 2006. His higher education includes M.Tech. (2009) and Ph. D. (2018), both from IIT Guwahati. He also spent some time with Processor Systems India Private Limited, Bangalore.

Dr. Sandeep is currently involved in teaching/research subjects related to signal processing.

Publications

Publication Type: Journal Article

Year of Publication Title

2019

Sandeep P. and Jacob, T., “Joint Color Space GMMs for CFA Demosaicking”, IEEE Signal Processing Letters, vol. 26, no. 2, pp. 232-236, 2019.[Abstract]


We propose a patch-based algorithm for demosaicking a mosaicked color image produced by color filter arrays commonly used in acquiring color images. The proposed algorithm exploits a joint color space Gaussian mixture model (JCS-GMM) prior for jointly characterizing the patches from red, green, and blue channels of a color image. The inter channel correlations captured by the covariance matrices of Gaussian models are exploited to estimate the pixel values missing in the mosaicked image. The proposed JCS-GMM demosaicking algorithm can be seen as the GMM analogue of the Color-KSVD algorithm, which has produced impressive results in color image denoising and demosaicking. We demonstrate that our proposed algorithm achieves superior performance in the case of Kodak and Laurent Condat's databases, and competitive performance in the case of IMAX database, when compared with state-of-the-art demosaicking algorithms.

More »»

2016

Sandeep P. and Jacob, T., “Single Image Super-Resolution Using a Joint GMM Method”, IEEE Transactions on Image Processing, vol. 25, no. 9, pp. 4233-4244, 2016.[Abstract]


Single image super-resolution (SR) algorithms based on joint dictionaries and sparse representations of image patches have received significant attention in the literature and deliver the state-of-the-art results. Recently, Gaussian mixture models (GMMs) have emerged as favored prior for natural image patches in various image restoration problems. In this paper, we approach the single image SR problem by using a joint GMM learnt from concatenated vectors of high and low resolution patches sampled from a large database of pairs of high resolution and the corresponding low resolution images. Covariance matrices of the learnt Gaussian models capture the inherent correlations between high and low resolution patches, which are utilized for inferring high resolution patches from given low resolution patches. The proposed joint GMM method can be interpreted as the GMM analogue of joint dictionary-based algorithms for single image SR. We study the performance of the proposed joint GMM method by comparing with various competing algorithms for single image SR. Our experiments on various natural images demonstrate the competitive performance obtained by the proposed method at low computational cost.

More »»

Publication Type: Conference Proceedings

Year of Publication Title

2014

Sandeep P. and Jacob, T., “Supervised dictionary learning for signals from union of subspaces”, 2014 International Conference on Signal Processing and Communications (SPCOM), vol. 1, No: 5. pp. 22-25, 2014.[Abstract]


Dictionary learning algorithms are used to train an overcomplete dictionary from a set of signal examples such that the learnt dictionary provides sparse representations for a class of signals from which the training examples are sampled. In this work, we consider a specific class of signals, i.e., signals which belong to a union of subspaces, and we propose a dictionary learning algorithm for such type of signals by extending the popular K-SVD algorithm. Apart from the traditional sparsity model, we also incorporate the union of subspaces model into the dictionary learning algorithm. Various experiments using synthetic and real data demonstrate that the proposed algorithm recovers a dictionary which is closer to the underlying unknown dictionary than the one obtained from a simple K-SVD algorithm which do not make use of the additional structure contained in the signal examples.

More »»

2013

Sandeep P. and Jacob, T., “Image restoration from multiple copies: A GMM based method”, IEEE International Conference on Acoustics, Speech and Signal Processing 2013 (ICASSP) , vol. 1593, No. 1597. IEEE, Vancouver, BC, Canada, pp. 26-31, 2013.[Abstract]


Recovery of original images from degraded and noisy observations is considered an important task in image processing. Recently, a Piece-wise Linear Estimator (PLE) was proposed for image recovery by using Gaussian Mixture Model (GMM) as a prior for image patches. Despite having much lesser computational requirements, this method yields comparable or better results when compared with the widely used sparse representation techniques for image restoration. In many situations, we might have access to multiple degraded copies of the same image, and would like to exploit the correlation among them for better image recovery. In this work, we extend the GMM based method to the multiple observations scenario, where we estimate the original image by utilizing the collective information available from all degraded copies.

More »»