mypic

Amol Mahurkar


email id: amolgmahurkar [AT] gmail.com



Short Bio: Amol Mahurkar's current research focus is on Compressive Sensing and Sparsity-based techniques with applications to Medical Imaging. His other research interests include Inverse Problems, Computational Sensing/Imaging, Signal Processing. More generally, he finds pleasure to delve deeper by investigating a mathematical function's interesting properties. He has an undergrad in ECE from Pune University, now known as Savitribai Phule Pune University.


Full bio on about me page.

Home Research Interests Education Publications Projects Codes CV Links About me

Research Interests

Education

Bachelor of Engineering, Electronics and Telecommunication
Pune University, India

[Find relevant coursework here from MOOCs such as Cousera]

Publications

[Scholar] [dblp]

Conferences

"SAMIR: Sparsity Amplified Iteratively-Reweighted Beamforming for High-Resolution Ultrasound Imaging"
Amol G Mahurkar*, Praveen Kumar Pokala*, Chandra Sekhar Seelamantula.
In the Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Oral, 2019. [Link to PDF]

"FirmNet: A Sparsity Amplified Deep Network for Solving Linear Inverse Problems"
Amol G Mahurkar*, Praveen Kumar Pokala*, Chandra Sekhar Seelamantula.
In the Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Oral, 2019. [Link to PDF]

"Iteratively-Reweighted Beamforming for High-Resolution Ultrasound Imaging"
Amol G Mahurkar, Praveen Kumar Pokala, Chandra Sekhar Seelamantula.
In the Proceedings of IEEE International Symposium on Biomedical Imaging (ISBI), 2019. [Link to PDF]

"Integrated Approach to Handwritten Character Recognition using ANN and Its Implementation on ARM"
Ganesh Rakate, Amol Mahurkar.
In the Proceedings of IEEE ICCIC 2012. [Link to PDF] [Demo Video] [More info] [Received Best Undergrad Project Award]

Posters/Abstracts

"Selective Visualization of Anomalies in Fundus Images via Sparse and Low Rank Decomposition"
Amol Mahurkar, Ameya Joshi, Naren Nallapareddy, Pradyumna Reddy, Micha Feigin, Achuta Kadambi, and Ramesh Raskar.
In ACM SIGGRAPH 2014 Posters (SIGGRAPH '14). [Link to PDF]

Book Contributions

"Chapter 1 - 3D Depth Cameras in Vision: Benefits and Limitations of the Hardware"
Achuta Kadambi, Ayush Bhandari, and Ramesh Raskar.
In Computer Vision and Machine Learning with RGB-D Sensors 2014, L Shao, J Han, P Kohli, Z Zhang. Springer International. [Link to PDF]

Projects

Regarding more recent projects:

Amol has worked on multiple research problems with various researchers around the world. Many hunches proved wrong, some succeeded.
Those (few) anomalies in his research can be found in Publications section.

  1. Vessel Extraction from retinal images using matched filter (as a part of MIT's REDX Initiative)
  2. Action Recognition on Kinect data (Final PA of CS228 Stanford)
  3. Classifying STL-10 images using Convolutional Neural Network
  4. Learning features with Sparse Auto-encoders
  1. Bayes Nets for Genetic Inheritance
  2. Markov Networks for OCR
  3. Exact Inference
  4. Approximate Inference
  5. Decision Making in assessing tests for Arrhythmogenic Right Ventricular Dysplasia (ARVD)
  6. CRF Learning for OCR
  7. Action Recognition on Kinect data
  1. Intelligent e-book (Electronic Notebook with Handwriting Recognition) [Received Best Undergrad Project Award]
  2. Auto-tune PID controller (Self Learning PID Controller)
  3. Checkpoint Assessment system for Line Following Robots
  4. Study of Vector Control Algorithm and Inverter design for BLDC Motor, V/f control Algorithm
  5. ROBOCON days
  6. My first line follower

Codes

Available on GitHub, MATLAB Central

  1. Implementation of Structured Least Squares with Bounded Data Uncertainties for System Identification [Link to Paper]
  2. Intelligent ebook (Electronic Notebook with Handwriting Recognition)
  3. Auto-tune PID controller (Self Learning PID Controller)
  4. UFLDL Sparse Auto-encoder
  5. UFLDL PCA 2-D (Unsupervised Feature Learning and Deep Learning)
  6. UFLDL PCA and ZCA Whitening
  7. UFLDL Softmax Regression
  8. UFLDL Self Taught Learning
  9. UFLDL Stacked Auto-encoder
  10. UFLDL Linear Decoders
  11. UFLDL Convolutional Neural Networks (CNN)
  12. Antenna Arrays in 3-D (Broadside, Endfire, Binomial)
  13. Antenna Arrays in 2-D (Broadside, Endfire, Binomial, Dolph-Chebyshev)
  14. Error Correcting Algorithms (Viterbi, Trellis and Cyclic Codec)
  15. and many more ...



* Drawing found on the top-right of the page was inspired from EE364a: Convex Optimization class by Prof. Stephen Boyd. Code available on request.


Theme inspired by Prof. Emmanuel Candes's webpage


Hosted on GitHub