Anurag Das

Anurag Das

PhD student in Computer Engineering at Texas A&M Univeristy, College Station

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

Blog Post number 4

less than 1 minute read

Published:

This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.

Blog Post number 3

less than 1 minute read

Published:

This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.

Blog Post number 2

less than 1 minute read

Published:

This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.

Blog Post number 1

less than 1 minute read

Published:

This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.

portfolio

publications

Image splicing detection using Gaussian or defocus blur

Published in 5th International Conference on Communication and Signal Processing(ICCSP), 2016

The availabilty of Image manipulation softwares such as Photoshop, has made it easier to forge images than ever before. In this work, we tackle a particular kind of image forgery called Image Splicing. In Image Splicing, an artificial region may be introduced in an image so as to alter its content. We address Image Splicing in natural images with inherent Gaussian blur. We hypothesize that the forged region introduced may have standard deviation different from the rest of the image or may have a different type of blur altogether. We aim to expose such imbalances in blur by first de-blurring the images and using ringing effects. The ringing effects, provide useful cues as to potential forgeries.

Download here

Copy-Move Forgery Detection using ASIFT

Published in India International Conference of Information Processing(IICIP), 2018

In this paper, we identify affine transform based Copy-Move forgeries in images. Copy-Move forgery is a type of image forgery where a portion of an image is pasted onto the same image so as alter its content. We compare the performance of ASIFT to that of SIFT in identifying such affine transformed copy-move forgeries. We show ASIFT outperforms SIFT in all of the test images, producing much larger number of keypoints per image and accurately identifying the forged regions.

Download here

Classification between story-telling and poem recitaiton using head gestures of the talker

Published in 12th International Conference on Signal Processing and Communication(SPCOM), 2018

In this work, we show that head gestures while reciting poems(rhythmic speech) have more periodic structure than while narrating stories. We use a dataset of 10 subjects reciting 20 poems and a seperate set of 20 subjects, each narrating 5 stories. To show the periodicity of head gestures, we use a measure based on the autocorrelation of the input signal. Using the information of peaks in the autocorrelation of an input signal, we achieve a highest periodicity of 0.489 in case of poems and a highest periodicity of 0.347 in case of stories. We further perform a classification task to classify between spontaneous speech and rhythmic speech. We show that head gesture features perform comparably well to the acoustic features(MFCCs), with accuracies 89% and 96% approximately. We further show that combination of head gestures and acoustic features outperform the acoustic features.

Download here

Understanding the effect of voice quality and accent on talker similarity

Published in Interspeech, 2020

This paper presents a methodology to study the role of non-native accents on talker recognition by humans. The methodology combines a state-of-the-art accent-conversion system to resynthesize the voice of a speaker with a different accent of her/his own, and a protocol for perceptual listening tests to measure the relative contribution of accent and voice quality on speaker similarity. Using a corpus of non-native and native speakers, we generated accent conversions in two different directions: non-native speakers with native accents, and native speakers with non-native accents. Then, we asked listeners to rate the similarity between 50 pairs of real or synthesized speakers. Using a linear mixed effects model, we find that (for our corpus) the effect of voice quality is five times as large as that of non-native accent, and that the effect goes away when speakers share the same (native) accent. We discuss the potential significance of this work in earwitness identification and sociophonetics.

Download here

A sparse coding approach to automatic diet monitoring with continuous glucose monitors

Published in ICASSP, 2021

Measuring dietary intake is a major challenge in the management of chronic diseases. Current methods rely on self-report measures, which are cumbersome to obtain and often unreliable. This article presents an approach to estimate dietary intake automatically by analyzing the post-prandial glucose response (PPGR) of a meal, as measured with continuous glucose monitors. In particular, we propose a sparse-coding technique that can be used to estimate the amounts of macronutrients (carbohydrates, protein, fat) in a meal from the meal’s PPGR. We use Lasso regularization to represent the PPGR of a new meal as a sparse combination of PPGRs in a dictionary, then combine the sparse weights with the macronutrient amounts in the dictionary’s meals to estimate the macronutrients in the new meal. We evaluate the approach on a dataset containing nine standardized meals and their corresponding PPGRs, consumed by fifteen participants. The proposed technique consistently outperforms two baseline systems based on ridge regression and nearest-neighbors, in terms of correlation and normalized root mean square error of the predictions.

Download here

talks

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.