My Projects

Practical Transfer Learning for Intelligent Robot Programming

Internship at Carnegie Mellon University with Prof. David Touretzky

Clustering white-matter fiber tracts of diffusion MRI

Internship at University of Melbourne with Prof. Kotagiri Rao

Simulation of transparent objects for estimating 6DOF pose

Internship at TU Wien with Prof. Markus Vincze

Accent recognition

Extracted and refined the Long and Short Term features from the audio data using PCA and HLDA. Improved the Accent classification by combining phonetic vowels with acoustic features and trained the model using a combination of Deep Neural Networks and Recurrent Neural Networks

Image Reconstruction with MRI technology

Awarded URA-01 for this notable contribution

Image classification with Liquid State Machine for Neuromorphic computing

Bachelors Thesis under the guidance of Prof. Udayan Ganguly

Innovation Cell

Contributed to control and hardware design of autonomous quadcopters in a GPS-denied environment. Implemented communication between on-board processor Intel i5 NUC, offboard computer and Pixhawk for transfer of localization and IMU data using MAVLink communication protocol on ROS. Utilized LiDAR sensors and Stereo Vision camera to maintain the current position of the quadcopter. Investigated optimum PID parameters which enhanced flight stability and performance

Classification of satellite image patches via Convolutional Neural Networks using Transfer learning

CE712(Digital Image Processing Of Remotely Sensed Data) course project

Neural Style Transfer

Implemented the research paper ”A Neural Algorithm of Artistic Style” using TensorFlow for texture transfer algorithm, that constrains a texture synthesis method by feature representations.

Multi Cycle Processor Design

EE224 Course Project