BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//EE - ECPv5.10.0//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:EE
X-ORIGINAL-URL:https://ee.iisc.ac.in
X-WR-CALDESC:Events for EE
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
BEGIN:STANDARD
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
DTSTART:20220101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220408T150000
DTEND;TZID=Asia/Kolkata:20220410T010000
DTSTAMP:20260615T084544
CREATED:20220325T033019Z
LAST-MODIFIED:20220328T230746Z
UID:239680-1649430000-1649552400@ee.iisc.ac.in
SUMMARY:EECS Resesarch Students Symposium 2022
DESCRIPTION:Click on the image to visit symposium website. Click here for the poster
URL:https://ee.iisc.ac.in/event/eecs-resesarch-students-symposium-2022/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220418T133000
DTEND;TZID=Asia/Kolkata:20220422T223000
DTSTAMP:20260615T084544
CREATED:20220405T095254Z
LAST-MODIFIED:20220406T004010Z
UID:239689-1650288600-1650666600@ee.iisc.ac.in
SUMMARY:Information for M.Tech aspirants in Electrical Engineering
DESCRIPTION:Department of Electrical Engineering \nIndian Institute of Science\, Bangalore \nImportant information to the applicants called for interview \nDear Applicant\, \nThis page is relevant to you only if you had applied for admission to M Tech EE programme and have been invited for an interview at the Department of Electrical Engineering\, IISc in April 2022. \nBased on your performance in GATE\, you have been shortlisted and invited to appear for a technical interview offline. There will not be any test. The final selection will be based on the performances in GATE and interviews. \nPlease note the following information: \nInterview will be held during 18 to 22 April 2022. So\, kindly adhere to the date(s) and time allotted for your interview. \nParticipation in interview is mandatory to be eligible for selection process. \nKindly carry your interview call letter from Academic section\, ID proof\, photostat copies of certificates\, Category Certificate\, National Qualifying exam Score card /Certificate and mark statements from 10th Std onwards for verification \nYou are requested to follow the Covid-19 related guidelines issued by the government \nWith our very best wishes\, \nChairman\, Department of Electrical Engineering \n  \nFor any queries mail to office.ee@iisc.ac.in. Or call at 22932361/3170
URL:https://ee.iisc.ac.in/event/information-for-m-tech-aspirants-in-electrical-engineering/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220418T163000
DTEND;TZID=Asia/Kolkata:20220418T173000
DTSTAMP:20260615T084544
CREATED:20220418T001523Z
LAST-MODIFIED:20220418T002022Z
UID:239695-1650299400-1650303000@ee.iisc.ac.in
SUMMARY:MTech Research Thesis Defense of Mr. Jaswanth Reddy Katthi @ 11am
DESCRIPTION:Location : Electrical Engineering\, MMCR (C241)\, Online via Teams (if network connection allows) https://tinyurl.com/2p8exxys \nTitle : Deep Learning Methods for Audio-EEG Analysis \nAbstract : The perception of speech and audio is one of the defining features of humans. Much of the brain’s underlying processes\, as we perceive acoustic signals\, are unknown\, and significant research efforts are needed to unravel them. The non-invasive recordings capturing the brain activations like electroencephalogram (EEG) and magnetoencephalogram (MEG) are commonly deployed to capture the brain responses to auditory stimuli. But these non-invasive techniques capture artifacts and noise that are not related to the stimuli\, which distort the downstream stimulus-response analysis.  The current state-of-art models used for normalization and pre-processing of EEG data utilize the linear canonical correlation analysis (CCA) or the temporal response function (TRF) based approach. However\, these methods assume a simplistic linear relationship between the audio features and the EEG responses and therefore\, may not alleviate the recording artifacts and interfering signals in EEG. This talk proposes novel methods using deep learning advances to improve the audio-EEG analysis. \nWe propose a deep learning framework for audio-EEG analysis in intra-subject and inter-subject settings. The deep learning based intra-subject analysis methods are trained with a Pearson correlation-based cost function between the stimuli and EEG responses. This model allows the transformation of the audio and EEG features in a common sub-space that is maximally correlated. The correlation-based cost function can be optimized with the learnable parameters of the model trained using standard gradient-descent based methods. This model is referred to as the deep CCA (DCCA) model. Several experiments\, performed on the EEG data recorded on subjects listening to naturalistic speech and music stimuli\, show that the deep methods obtain improved representations than the linear methods\, thereby resulting in statistically significant improvements in correlation values. \nFurther\, we propose a neural network model with shared encoders that align the EEG responses from multiple subjects listening to the same audio stimuli. This inter-subject model boosts the signals common across the subjects related to the stimuli and suppresses the subject-specific artifacts. This model is referred to as the deep multi-way canonical correlation analysis (DMCCA). The combination of inter-subject analysis using DMCCA and intra-subject analysis using DCCA is shown to provide the best stimulus-response in audio-EEG experiments. \nFinally\, the talk will discuss about an ambitious experiment\, where we attempted to recreate acoustic signal directly from EEG responses. While the audio is not fully recoverable\, the parts of the signal that can be recovered from the non-invasive EEG recordings throws light into the characteristics of audio captured in the EEG data.
URL:https://ee.iisc.ac.in/event/mtech-research-thesis-defense-of-mr-jaswanth-reddy-katthi-11am/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220420T203000
DTEND;TZID=Asia/Kolkata:20220420T213000
DTSTAMP:20260615T084544
CREATED:20220418T002901Z
LAST-MODIFIED:20220418T003005Z
UID:239705-1650486600-1650490200@ee.iisc.ac.in
SUMMARY:Seminar by Prof. Sanjib Kumar Panda @ 3pm
DESCRIPTION:Title: Single-phase inverter control techniques for interfacing renewable energy sources with micro-grid – Parallel connected inverter topologies with active and reactive power flow control along with grid current shaping \nTime: 20 April 2022\, 3:00 pm \nVenue: MMCR EE \nAbstract: Renewable energy sources (RESs) have been receiving significant attention recently worldwide as a sustainable alternative type of energy supply in the energy mix. Inverters are being used to convert the dc voltage into ac voltage before being injected into the grid or isolated loads. In this presentation\, a novel current control technique is proposed to control both active and reactive power flow from a renewable energy source feeding a micro-grid system through a single-phase parallel connected inverter. The parallel-connected inverter ensures active and reactive power flow from the grid with low current THD even in the presence of non-linear load. A p-q theory-based approach is used to find the reference current of the parallel-connected converter to ensure desired operating conditions at the grid terminal. The proposed current controller is simple to implement and gives superior performance over the conventional current controllers such as rotating frame PI controller or stationary frame Proportional Resonant (PR) controller. The stability of the proposed controller is ensured by the direct Lyapunov method. A new technique based on the Spatial Repetitive Controller (SRC) is also proposed to improve the performance of the current controller by estimating the grid and other periodic disturbances. Detailed experimental results are presented to show the efficacy of the proposed current control scheme along with the proposed non-linear controller to control the active and reactive power flow in a single-phase micro-grid under different operating conditions \nSpeaker’s Bio: Sanjib Kumar Panda (Student 98’\, Member ’92\, SM 00’\, F 21’) received a Bachelor of Engineering Degree with 1st Class Honours from Sardar Vallabhabhai Regional College of Engineering and Technology\, Surat\, India\, in 1983. He was awarded the Gold Medal for securing the highest marks not only amongst the B. Engg. (Electrical) but also for securing the highest marks amongst all the B. Engg. (Civil\, Mechanical and Electrical) students. He also earned a Masters of Technology Degree from the Institute of Technology\, Banaras Hindu University\, Varanasi\, India in 1987. He was awarded the Gold Medal for securing the highest marks amongst all the M. Tech. (Electrical) students. Subsequently\, he earned a PhD. Degree from the University of Cambridge\, U.K.\, in 1991. He was awarded the Nehru Cambridge Fellowship and Overseas Research Studentship from the Cambridge Commonwealth Trust for Cambridge University for his PhD studies\, 1987-1991. \nHe joined the Department of Electrical and Computer Engineering at the National University of Singapore as a Lecturer in 1992. He is currently serving as an Associate Professor and Director of the Power & Energy Research Area. He has served as Director (Education) at the Design Technology Institute\, a joint-venture between NUS and TU/e\, The Netherlands and funded by EDB\, Singapore. He has served as the Group Head of the Drives\, Power and Control Group from 2007-2009. He was appointed as Area Director\, Power & Energy Research Group of the Department of Electrical & Computer Engineering at NUS on 1st January 2010 and serving in this position till date. \nDr. Panda has won the Annual Teaching Excellence Award at the National University of Singapore in 2004 and 2009. Besides these two University Level Awards\, he has also been awarded several Teaching Awards at the Faculty of Engineering and at the Department of Electrical and Computer Engineering Department consistently since the year 2000. \nDr. Panda has carried out extensive research in various areas of control of electric drives and power electronic converters. He has co-authored 1 book\, several book chapters\, published more than 450 papers in international refereed journals and conferences and holds 6 patents to his credit. Dr. Panda has an h-index of 44 and has citations almost close to 10\,000. He has received research funding to the tune of S$25mil over the past 15 years or so. Dr. Panda is the co-founder of three start-up companies: (1) ENBED Pte. Ltd.\, (2) REPMIX Pte. Ltd. and (3) SPCSCAN Pte. Ltd. along with his PhD students and research staff. His current research interests are in high-performance control of motor drives\, control of distributed renewable energy sources and their integration with grid\, condition monitoring\, preventive and predictive maintenance. \nDr. Panda has been an active member of the IEEE. He has served in various capacities as Chapter Officer in the IEEE Singapore Section’s Joint Power Electronics and Industry Applications Society Chapter since 1994 till date. He has served in various capacities in the IEEE Singapore Section during the period 2000-2004 and served as the Section Chairman during the period 2004. He was the recipient of the IEEE 3rd Millennium Medal. He was the Organizing Chairman for the International IEEE Power Electronics and Drives Systems Conference (PEDS) in 2003. Dr. Panda served as the founding Chairman for the International Conference on Sustainable Energy Technologies (ICSET) in 2008. He was awarded the Best Volunteer Award by the IEEE Singapore Section in 2006. He was awarded the Best Volunteer Award by IEEE R-10 in 2014. Since 2012\, Dr. Panda has been a volunteer serving in the Membership and Chapter Development for the IEEE PELS and presently serving as R-10 Coordinator. The IEEE PELS has seen the consistent membership growth rate of more than 15% for the R-10. Dr. Panda also proposed the Regional Distinguished Lecture (RDL) Program for the IEEE PELS and the initiated as a part of the R-10 RDL Speakers to be implemented in June 2020. Dr. Panda is the Organizing Chair for the IEEE ECCE-ASIA 24-27th May\, 2021 to be held at Singapore. Dr. Panda has also served as a Member in the Program Committee in the IEEE Section Congress 2014 at Amsterdam\, The Netherlands. He also presented in the IEEE Section Congress 2017 at Sydney\, Australia. Dr. Panda has been serving as an Associate Editor of the IEEE Transactions in Power Electronics\, the Journal of Emerging and Selected Topics in Power Electronics since 2012 till date. Dr. Panda has been elevated to the IEEE Fellowship w.e.f. form 1st Jan. 2021. Dr. Panda is the IEEE PELS DL for the period 1st Jan. 2022 – 31st December 2023. \nAll are welcome.
URL:https://ee.iisc.ac.in/event/seminar-by-prof-sanjib-kumar-panda-3pm/
LOCATION:EE\, MMCR
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220420T203000
DTEND;TZID=Asia/Kolkata:20220420T213000
DTSTAMP:20260615T084544
CREATED:20220426T025140Z
LAST-MODIFIED:20220426T025140Z
UID:239714-1650486600-1650490200@ee.iisc.ac.in
SUMMARY:Seminar by Prof. Sanjib Kumar Panda
DESCRIPTION:meeting link: Teams meeting link \nTitle: Single-phase inverter control techniques for interfacing renewable energy sources with micro-grid – Parallel connected inverter topologies with active and reactive power flow control along with grid current shaping \n\n\n\n Speaker: Professor Sanjib Kumar Panda \n Time: 20 April 2022\, 3:00 pm \n Venue: MMCR EE \n Abstract: Renewable energy sources (RESs) have been receiving significant attention recently worldwide as a sustainable alternative type of energy supply in the energy mix. Inverters are being used to convert the dc voltage into ac voltage before being injected into the grid or isolated loads. In this presentation\, a novel current control technique is proposed to control both active and reactive power flow from a renewable energy source feeding a micro-grid system through a single-phase parallel connected inverter. The parallel-connected inverter ensures active and reactive power flow from the grid with low current THD even in the presence of non-linear load. A p-q theory-based approach is used to find the reference current of the parallel-connected converter to ensure desired operating conditions at the grid terminal. The proposed current controller is simple to implement and gives superior performance over the conventional current controllers such as rotating frame PI controller or stationary frame Proportional Resonant (PR) controller. The stability of the proposed controller is ensured by the direct Lyapunov method. A new technique based on the Spatial Repetitive Controller (SRC) is also proposed to improve the performance of the current controller by estimating the grid and other periodic disturbances. Detailed experimental results are presented to show the efficacy of the proposed current control scheme along with the proposed non-linear controller to control the active and reactive power flow in a single-phase micro-grid under different operating conditions \nSpeaker’s Bio: Sanjib Kumar Panda (Student 98’\, Member ’92\, SM 00’\, F 21’) received a Bachelor of Engineering Degree with 1st Class Honours from Sardar Vallabhabhai Regional College of Engineering and Technology\, Surat\, India\, in 1983. He was awarded the Gold Medal for securing the highest marks not only amongst the B. Engg. (Electrical) but also for securing the highest marks amongst all the B. Engg. (Civil\, Mechanical and Electrical) students. He also earned a Masters of Technology Degree from the Institute of Technology\, Banaras Hindu University\, Varanasi\, India in 1987. He was awarded the Gold Medal for securing the highest marks amongst all the M. Tech. (Electrical) students. Subsequently\, he earned a PhD. Degree from the University of Cambridge\, U.K.\, in 1991. He was awarded the Nehru Cambridge Fellowship and Overseas Research Studentship from the Cambridge Commonwealth Trust for Cambridge University for his PhD studies\, 1987-1991. \nHe joined the Department of Electrical and Computer Engineering at the National University of Singapore as a Lecturer in 1992. He is currently serving as an Associate Professor and Director of the Power & Energy Research Area. He has served as Director (Education) at the Design Technology Institute\, a joint-venture between NUS and TU/e\, The Netherlands and funded by EDB\, Singapore. He has served as the Group Head of the Drives\, Power and Control Group from 2007-2009. He was appointed as Area Director\, Power & Energy Research Group of the Department of Electrical & Computer Engineering at NUS on 1st January 2010 and serving in this position till date. \nDr. Panda has won the Annual Teaching Excellence Award at the National University of Singapore in 2004 and 2009. Besides these two University Level Awards\, he has also been awarded several Teaching Awards at the Faculty of Engineering and at the Department of Electrical and Computer Engineering Department consistently since the year 2000. \nDr. Panda has carried out extensive research in various areas of control of electric drives and power electronic converters. He has co-authored 1 book\, several book chapters\, published more than 450 papers in international refereed journals and conferences and holds 6 patents to his credit. Dr. Panda has an h-index of 44 and has citations almost close to 10\,000. He has received research funding to the tune of S$25mil over the past 15 years or so. Dr. Panda is the co-founder of three start-up companies: (1) ENBED Pte. Ltd.\, (2) REPMIX Pte. Ltd. and (3) SPCSCAN Pte. Ltd. along with his PhD students and research staff. His current research interests are in high-performance control of motor drives\, control of distributed renewable energy sources and their integration with grid\, condition monitoring\, preventive and predictive maintenance. \nDr. Panda has been an active member of the IEEE. He has served in various capacities as Chapter Officer in the IEEE Singapore Section’s Joint Power Electronics and Industry Applications Society Chapter since 1994 till date. He has served in various capacities in the IEEE Singapore Section during the period 2000-2004 and served as the Section Chairman during the period 2004. He was the recipient of the IEEE 3rd Millennium Medal. He was the Organizing Chairman for the International IEEE Power Electronics and Drives Systems Conference (PEDS) in 2003. Dr. Panda served as the founding Chairman for the International Conference on Sustainable Energy Technologies (ICSET) in 2008. He was awarded the Best Volunteer Award by the IEEE Singapore Section in 2006. He was awarded the Best Volunteer Award by IEEE R-10 in 2014. Since 2012\, Dr. Panda has been a volunteer serving in the Membership and Chapter Development for the IEEE PELS and presently serving as R-10 Coordinator. The IEEE PELS has seen the consistent membership growth rate of more than 15% for the R-10. Dr. Panda also proposed the Regional Distinguished Lecture (RDL) Program for the IEEE PELS and the initiated as a part of the R-10 RDL Speakers to be implemented in June 2020. Dr. Panda is the Organizing Chair for the IEEE ECCE-ASIA 24-27th May\, 2021 to be held at Singapore. Dr. Panda has also served as a Member in the Program Committee in the IEEE Section Congress 2014 at Amsterdam\, The Netherlands. He also presented in the IEEE Section Congress 2017 at Sydney\, Australia. Dr. Panda has been serving as an Associate Editor of the IEEE Transactions in Power Electronics\, the Journal of Emerging and Selected Topics in Power Electronics since 2012 till date. Dr. Panda has been elevated to the IEEE Fellowship w.e.f. form 1st Jan. 2021. Dr. Panda is the IEEE PELS DL for the period 1st Jan. 2022 – 31st December 2023. \n All are welcome.
URL:https://ee.iisc.ac.in/event/seminar-by-prof-sanjib-kumar-panda/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220420T230000
DTEND;TZID=Asia/Kolkata:20220421T000000
DTSTAMP:20260615T084544
CREATED:20220411T231422Z
LAST-MODIFIED:20220411T231422Z
UID:239692-1650495600-1650499200@ee.iisc.ac.in
SUMMARY:Seminar by Prof. Sairaj Dhople @ 5.30pm
DESCRIPTION:Title: Power-system Modeling & Control for the Era of Inverter-based Resources \nSpeaker: Prof. Sairaj Dhople\, Electrical & Computer Engineering\, University of Minnesota\nTime: 20 April 2022\, 5:30 pm\nVenue: MMCR EE / Hybrid mode \nhttps://teams.microsoft.com/l/meetup-join/19%3ameeting_YWJjZTQyZTAtYTFjNi00YTVkLWE0NjktYmJkZDMzNjI4ZDFm%40thread.v2/0?context=%7b%22Tid%22%3a%226f15cd97-f6a7-41e3-b2c5-ad4193976476%22%2c%22Oid%22%3a%220c8fe27d-52c2-4e77-a28f-a759bd113fae%22%7d \nAbstract: Power networks all over the world are experiencing dramatic upheaval in compositional form and anticipated functionality. With retirement of fossil-fuel-driven synchronous generators\, integration of renewable energy\, and adoption of electrified transportation\, there is a pronounced change in the energy-conversion interfaces that form the backbone of the grid. Particularly\, energy processing in future grids will be dominantly handled by semiconductor-based power-electronics circuits termed inverter-based resources (IBRs). This talk will provide snapshots of how classical power-system modeling problems can (and will have to) be revised to accomodate these emerging technologies. In particular\, we will present insights on synchronization of IBRs with a variety of control methods\, provide a system-theoretic solution to normalizing dynamic models of diverse grid assets\, and overview a time-domain network-reduction method for large-scale electrical networks. Each topic will be presented with an effort to acknowledge the rich history of personalities\, methods\, and venues relevant to power engineering over the 20th century. Finally\, we will discuss how partnerships and collaboration across academia\, industry (system operators\, utilities\, manufacturers)\, and national labs will be critical to facilitate large-scale integration with performance guarantees. \nBio: Sairaj Dhople received the B.S.\, M.S.\, and Ph.D. degrees in electrical engineering from the University of Illinois at Urbana-Champaign\, Urbana\, IL\, USA\, in 2007\, 2009\, and 2012\, respectively. He is currently serving as Associate Professor with the Department of Electrical and Computer Engineering at the University of Minnesota. His research interests include modeling\, analysis\, and control of power electronics and power systems with a focus on renewable integration. He is the recipient of the National Science Foundation CAREER Award in 2015\, the Outstanding Young Engineer Award from the IEEE Power and Energy Society in 2019\, and the IEEE Power and Energy Society Prize Paper Award in 2021.
URL:https://ee.iisc.ac.in/event/seminar-by-prof-sairaj-dhople-5-30pm/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220421T163000
DTEND;TZID=Asia/Kolkata:20220421T173000
DTSTAMP:20260615T084544
CREATED:20220426T022712Z
LAST-MODIFIED:20220426T022712Z
UID:239712-1650558600-1650562200@ee.iisc.ac.in
SUMMARY:MTech (Rresearch) Colloquium
DESCRIPTION:Name of the Student:      Ahmad Arfeen \nGuide:                             Prof. Soma Biswas \nDate And Day:                21st April\, 2022\, Thursday \nTime:                               11:00 am \nVenue:                              EE\, MMCR \nTitle : Data Efficient Domain Generalization \nAbstract: For the task of image classification\, in general\, the test data is assumed to come from the same distribution as the training data. But this may not always hold in real-life scenarios. For example\, in night-time surveillance\, we may need to classify images captured using NIR cameras\, but the available model has been trained on RGB images. Domain generalization (DG) addresses the problem of generalizing classification performance across any unknown domain\, by leveraging training samples from multiple source domains. In this thesis\, we address two challenging scenarios for the DG task. Currently\, the training process of majority of the state-of-the-art DG-methods is dependent on a large amount of labeled data. This restricts the application of the models in many real-world scenarios\, where collecting and annotating a large dataset is an expensive and difficult task.  \nThus\, as the first contribution\, we address the problem of Semi-supervised Domain Generalization (SSDG)\, where the training set contains only a few labeled data\, in addition to a large number of unlabeled data from multiple domains. This is relatively unexplored in literature and poses a considerable challenge to the state-of-the-art DG models\, since their performance degrades under such condition. To address this scenario\, we propose a novel Selective Mixing and Voting Network (SMV-Net)\, which effectively extracts useful knowledge from the set of unlabeled training data\, available to the model. Specifically\, we propose a mixing strategy on selected unlabeled samples on which the model is confident about their predicted class labels to achieve a domain-invariant representation of the data\, which generalizes effectively across any unseen domain. Extensive experiments on two popular DG-datasets demonstrate the usefulness of the proposed framework.  \nThe second contribution of this thesis is a novel approach for the task of Zero-Shot Domain Generalization (ZSDG). This is very challenging since the query data can belong to an unseen class as well as unseen domain. For this task\, we address the challenge of class imbalance by learning class-specific classifier margins\, which not only maintain the semantic relationship of the classes in the embedding space\, but is also discriminative\, and thus improves the classification performance on the test data. Extensive experiments on multiple datasets justify the effectiveness of the proposed approach. \nALL ARE WELCOME
URL:https://ee.iisc.ac.in/event/mtech-rresearch-colloquium/
LOCATION:EE\, MMCR
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220422T213000
DTEND;TZID=Asia/Kolkata:20220422T223000
DTSTAMP:20260615T084544
CREATED:20220418T002346Z
LAST-MODIFIED:20220418T002455Z
UID:239701-1650663000-1650666600@ee.iisc.ac.in
SUMMARY:M. Tech. (Research) Thesis Colloquium of Anwesha Roy
DESCRIPTION:Title: Improved air-tissue boundary segmentation in real-time magnetic resonance imaging videos using speech articulator specific error criterion \nAbstract: Real-time Magnetic Resonance Imaging (rtMRI) is a tool used exhaustively in speech science and linguistics to understand the dynamics of the speech production process across languages and health conditions. rtMRI has two advantages over other methods which capture articulatory movement\, like X-ray\, Ultrasound and Electromagnetic articulography – it is non-invasive\, and it captures a complete view of the vocal tract including pharyngeal structures. The rtMRI video provides spatio-temporal information of speech articulatory movements\, which helps in modeling speech production. For this purpose\, a common step is to obtain the air-tissue boundary (ATB) segmentation in all frames of the rtMRI video. The accurate estimation of ATBs of the upper airway of the vocal tract is essential for many speech processing applications like speaker verification\, text-to-speech synthesis\, visual augmentation for synthesized articulatory videos\, and analysis of vocal tract movement. Thus\, it is necessary to have an accurate air-tissue boundary segmentation in every frame of the rtMRI videos. \nThe best performance in ATB segmentation of rtMRI videos in speech production\, in unseen subject conditions\, is known to be achieved by a 3-dimensional convolutional neural network (3D-CNN) model. In seen subject conditions\, both 3D-CNN and 2-dimensional deep convolutional encoder-decoder network (SegNet) show similar performance. However\, the evaluation of these models\, as well as other ATB segmentation techniques reported in literature\, has been done using Dynamic Time Warping (DTW) distance between the entire original and predicted boundaries or contours. Such an evaluation measure may not capture local errors in the predicted contour. Careful analysis of predicted contours reveals errors in regions like the velum part and tongue base section\, which are not captured in a global evaluation metric like DTW distance. In this thesis\, we automatically detect such errors and propose a novel correction scheme for them. We also propose two new evaluation metrics for ATB segmentation\, separately for each contour\, to explicitly capture errors in these contours. \nMoreover\, the state-of-the-art models use overall binary cross entropy as the loss function during model training. However\, such a global loss function does not give enough emphasis on regions which are more prone to errors. In this thesis\, together with global loss\, we explore the use of regional loss functions which focus on areas of the contours which have been analyzed as error prone in our analysis. Two different losses are considered in the regions around velum and tongue base – binary cross entropy (BCE) loss and dice loss. It is observed that dice-loss based models perform better than their BCE loss based counterparts.
URL:https://ee.iisc.ac.in/event/m-tech-research-thesis-colloquium-of-anwesha-roy/
LOCATION:EE\, MMCR
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20220428T163000
DTEND;TZID=Asia/Kolkata:20220428T173000
DTSTAMP:20260615T084544
CREATED:20220426T020023Z
LAST-MODIFIED:20220426T020102Z
UID:239710-1651163400-1651167000@ee.iisc.ac.in
SUMMARY:PhD Thesis Colloquium
DESCRIPTION:Name of Student: Ruturaj Gavaskar. \nGuide: Prof. Kunal Narayan ChaudhuryDate:  April 28\, Thursday.               Time: 11-12 am.Venue: MS Teams (online).Link: https://tinyurl.com/bdfardzz \nTitle:  On plug-and-play regularization using linear denoisers.Abstract:  The problem of inverting a given measurement model comes up in several  computational imaging applications. For example\, in CT and MRI\, we are  required to reconstruct a high-resolution image from incomplete noisy  measurements\, whereas in superresolution and deblurring\, we try to infer  the ground-truth from low-resolution or blurred images. While several  forms of regularization and associated optimization methods have been  proposed in the imaging literature of the last few decades\, the use of  denoisers (aka denoising priors) for image regularization is a  relatively recent phenomenon. This has partly been triggered by the  advances in image denoising in the last 20 years\, leading to the  development of powerful image denoisers. In this thesis\, we look at a  recent protocol called Plug-and-Play (PnP) method\, where powerful image  denoisers such as BM3D and DnCNN are deployed within iterative  algorithms for image regularization. Surprisingly\, the reconstructed  images are of high quality and competitive with state-of-the-art  methods. Following this\, researchers have tried explaining why plugging  a denoiser within an inversion algorithm should work in the first place\,  why it produces high-quality images\, and whether the final  reconstruction is optimal in some sense. We have tried answering some of  these questions in this thesis.At a high level\, the contributions of the thesis are as follows. Based  on the theory of proximal operators\, we prove that a PnP algorithm in  fact minimizes a convex objective function provided the plugged denoiser  belongs to a broad class L of linear filters. In particular\, L has a  simple characterization and includes kernel and GMM denoisers. That we  are able to characterize the reconstruction (for class L denoisers) as  the solution of a convex optimization problem helps in settling some of  the above questions. For example\, this allows us to establish iterate  convergence for PnP regularization. Obtaining such a guarantee for  complex nonlinear denoisers such as BM3D and neural denoisers is  nontrivial. As a more profound application\, we are able to provide  guarantees on signal recovery for the compressed sensing problem. More  precisely\, under certain verifiable assumptions\, we are able to prove  that a signal can be recovered exactly (resp. stably) with high  probability from random clean (resp. noisy) measurements using PnP  regularization. To the best of our knowledge\, this is the first such  result where the underlying assumptions are verifiable. We will present  and discuss these and other theoretical findings in greater detail  during the colloquium. We will also present numerical results to  validate our findings.
URL:https://ee.iisc.ac.in/event/phd-thesis-colloquium/
END:VEVENT
END:VCALENDAR