Hi I'm Mukesh B R. A Final Year Inforamtion Technology Student of National Institute of Technology Karnataka Surathkal.
I'm from Bangalore,Karnataka. My interests in Computer Science lies in Deep Learning and Algorithms.
In particular I'm interested in the field of computer vision. My hobbies include reading books and watching TV shows.
April 2020 - Present
Working on the problem of "Membrane Detection and Skeletonization of Electron Microscopy(EM) Images" of the brain using a derivative of the Inception-ResU-Nets with Squeeze-Excite characteristics. Achieved a Dice Coefficient Score of 0.93 and 0.91 for the detection of cell membranes and skeletonization of constituent cells respectively on the SNEMI3D dataset. The pipeline would lead to improvements to the standard automated segmentation algorithms. Also worked on the problem of correcting erroneous neuronal annotations, which is currently under submission at CVPR 2021.
Developed a statistical machine learning based algorithm to accurately simulate process of mixing chem icals, thereby providing operator assistance by predicting estimated time of completion. icals, thereby providing operator assistance by predicting estimated time of completion. The pipeline was designed to take into account dynamic variation of system parameters and operating conditions while obeying the laws of physics. Achieved a state of the art prediction with an error of 0.002 while predicting the system pH.
August 2020 - Present
Developed a distributed caching system to cache in-process data from a microservices oriented server. Achieved a speedup of around 20-25x while retrieving data from the database. Developed a two-step fallback mechanism to ensure eventual consistency of critical business logic data with re-publication using a publisher-subscriber messaging based interface to increase resiliency.
May 2020 - June 2020
During my summer break between the 4th annd 5th semesters, I interned at the prestigious Spectrum Lab situated at the Indian Institute of Science Bangalore. Under the guidance of Dr. Chandra Shekhar Seelamanthula I worked on Diabetic Retinopathy detection and grading. Using a variant of VGG+FCNN I was able to beat the state of the art in Referable Diabetic Retinopathy Grading with an AUC of 0.979.
July 2017 - Present
I'm curretly a Senior at NITK Surathkal Inforamtion Technology Department. My CGPA so far is 9.51 over the first 7 semesters.
June 2015 - May 2017
I studied class 11 and 12 from National Public School Rajajinagar,Bangalore. In my AISSCE exam I scored a 93.6%.
July 2008 - May 2015
I studied class 6 to 10 from Venkat International Public School(High) Rajajinagar,Bangalore. In my AISSE exam I scored CGPA of 10.
A concept or abstract can be truely understood by implementing their ideas by projects.
Here are few of my ongoing and Completed projects.
Developed a bio-metrics based authentication mechanism to authenticate users based on facial feature movements when uttering a passphrase. The proposed method achieves a state of the art Equal Error Rate of 0.023 on the MIRACL-VC1 dataset. Accepted for publication at the 35th AAAI Conference on Artificial Intelligence, Student Abstract and Poster Program 2021
Developed an autoencoder based dimensionaliy reduction algorithm enhanced with K-Means Clustering which also serves as an encryption mechanism ensuring security during data transmission. Accepted for publication in the Evolution in Computational Intelligence series, Springer Nature
This paper was accepted for presentation at the 3rd International Conference on Trends in Electronics and Informatics (ICOEI-2019) . This paper was an attempt to compare Deep Learning frameworks such as Keras and Torch. The metrics used were Train Times and Test Times while using CPU’s and GPU’s while trying fit standard model architectures like Alex Net on popular Datasets like the MINST and CIFAR-10.
An android app to detect the reach of various hurricanes affecting the world. This uses real time data collected by satellites and weather organizations from across the word. The data is then analyzed by a pre trained Random Forrest Regression model to detect the presence of a hurricane and its reach. It then warns users who are in this region and guides the users to a set of precautionary measures.
This paper was accepted for presentation at the 3rd International Conference on Trends in Electronics and Informatics (ICOEI-2019). This paper was an attempt to compare Deep Learning frameworks such as Keras and Torch. The metrics used were Train Times and Test Times while using CPU’s and GPU’s while trying fit standard model architectures like Alex Net on popular Datasets like the MINST and CIFAR-10.
Gender and Ethnicity Classification and Generation using Deep Learning in Heterogeneous Face Recognition with the help of CNN, GAN's and a simple web application.
A mobile app and a portal with a built in QR-Code Scanner to automate mess attendance and Management and make it more efficient
This research project was aimed at a heuristic sub-optimal solution to the Travelling Salesman Problem using the Ant colony Optimization (Swarm Heuristic) Technique.
A research project to compare newly discovered WAVL trees with other Balanced Binary Search Trees(BST’s) such as an AVl and Red-Black Tree. For details please Refer : Rank balanced BST
This research project was aimed at a heuristic sub-optimal solution to the Task Scheduling problem on a multi-processor environment using the Genetic Algorithm Technique.
I truely belive a student is characterized by both by their curricular and Extra Curriculars. Extra Curriculars build up a person's ability to socialize and work as a part of a group.
Incident is the annual cultural festival of the National Institute of Technology Karnataka (NITK) in Surathkal, Karnataka, India. The fest is a five-day-long event during the even semester of the NITK Calendar held in the month of March. The fest brings together students from different colleges all over India and is composed of several competitive events as well as pro-shows by big artists including Sunidhi Chauhan, Salim-Sulaiman, the duo of Vishal-Shekhar and Javed Ali. Incident is belived to be the second biggest cultural fest in India (after Mood Indigo) due to the sheer scale and number of events hosted.
We see a world where computing helps solve tomorrow’s problems – where we use our knowledge and skills to advance the profession and make a positive impact. ACM, the world's largest educational and scientific computing society, delivers resources that advance computing as a science and a profession. ACM provides the computing field's premier Digital Library and serves its members and the computing profession with leading-edge publications, conferences, and career resources. ACM NITK plans and organizes a wide range of events that include technical workshops, guest lectures, online events and various other competitions at NITK, throughout the year.
TEDxNITKSurathkal is meant to reach out to the students and faculty of NITK, Surathkal, an institute of engineering located in the Western Ghats in Karnataka, India. Being one of the premier institutes to study engineering in India, the college attracts some of the brightest minds in the country and has a vibrant and dynamic student community
In my 3rd Semester, Me and my teammates created an Mobile Application to help predict and detect hurricane windspeeds. This project won us the 2nd Place in the Microsoft Codefundo in our college. The source code is avaliable my Github Page.
In class 12 me and my teamamtes won the Bangalore Materials Quiz conductedd in IISC,Bengaluru. We won the quiz and got a chance to represent Karnataka in the BPMMQ quiz held at IGCAR, Kalpakkam.
Any Concerns and doubts about me please feel free to contact me
Phone: (720) 761-9416
mb4862@columbia.edu
mukeshbr.br309@gmail.com
1090 Saint Nicholas Avenue
Apartment 62
New York City,NY 10032