Prithviraj Sengupta

Prithviraj Sengupta

PhD Student at UIC

Publications

A Review of and Some Results for Ollivier–Ricci Network Curvature

Published in Mathematics, 2020

Characterizing topological properties and anomalous behaviors of higher-dimensional topological spaces via notions of curvatures is by now quite common in mainstream physics and mathematics, and it is therefore natural to try to extend these notions from the non-network domains in a suitable way to the network science domain. In this article we discuss one such extension, namely Ollivier’s discretization of Ricci curvature. We first motivate, define and illustrate the Ollivier–Ricci Curvature. In the next section we provide some “not-previously-published” bounds on the exact and approximate computation of the curvature measure. In the penultimate section we review a method based on the linear sketching technique for efficient approximate computation of the Ollivier–Ricci network curvature. Finally in the last section we provide concluding remarks with pointers for further reading.

Download here

Classification of Computer Generated and Natural Images based on Efficient Deep Convolutional Recurrent Attention Model

Published in CVPR Workshops, 2019

Most state-of-the-art techniques of distinguishing natural images and computer generated images based on handcrafted feature and Convolutional Neural Network require processing of the entire input image pixels uniformly. As a result, such techniques usually require extensive computation time and memory, that scale linearly with the size of the input image in terms of number of pixels. In this paper, we deploy an efficient Deep Convolutional Recurrent Attention model with relatively less number of parameters, to distinguish between natural and computer generated images. The proposed model uses a glimpse network to locally process a sequence of selected image regions; hence, the number of parameters and computation time can be controlled effectively. We also adopt a local-to-global strategy by training image patches and classifying full-sized images using the simple majority voting rule. The proposed approach achieves superior classification accuracy compared to recently proposed approaches based on deep learning.

Download here

Source anonymization of digital images: a counter–forensic attack on PRNU based source identification techniques

Published in Annual Conference on Digital Forensics, Security and Law, 2017

A lot of photographers and human rights advocates need to hide their identity while sharing their images on the internet. Hence, source-anonymization of digital images has become a critical issue in the present digital age. The current literature contains a few digital forensic techniques for” source-identification” of digital images, one of the most efficient of them being Photo-Response Non-Uniformity (PRNU) sensor noise pattern based source detection. PRNU noise pattern being unique to every digital camera, such techniques prove to be highly robust way of sourceidentification. In this paper, we propose a counter-forensic technique to mislead this PRNU sensor noise pattern based source-identification, by using a median filter to suppress PRNU noise in an image, iteratively. Our experimental results prove that the proposed method achieves considerably higher degree of source anonymity, measured as an inverse of Peak-to-Correlation Energy (PCE) ratio, as compared to the state-of-the-art.

Download here