Experience

 
 
 
 
 
January 2014 – June 2014
Mumbai, India

Research Intern

Bhabha Atomic Research Centre

Worked on compressing Magnetic Flux leakage signals collected by Instrumented Pipeline Inspection Gauage while monitoring oil/gas pipelines. Thesis
 
 
 
 
 
February 2012 – December 2013
Manipal, India

Head of Communication and Groundstation subsystem

Parikshit student satellite team

Responsibilities include:

  • Programmed ADF7021-N and CC1101 RF transcievers using MSP 430 for communication between satellite and groundstation
  • Designed RF PCB components used onboard Satelilte
  • Built a groundstation for tracking the satellites

Publications

We further propose a novel training procedure which starts out with a dense convolution but gradually evolves towards a grouped …

Nano Satellites generally weighing between 1 and 15 kg, serving the purpose of space research are popularly standardized as CubeSats. …

Talks

This is a list about the talks I have given this year at two events. One is Symposium on Research and other is CVIT Tech Talk. Symposium on research is an Annual Event conducted by CVIT, IIIT where students of CVIT will be selected to present their published research. I was given the opportunity to present my research Efficient Semantic Segmentation using Gradual Grouping. CVIT Tech Talk is a weekly meetup to discuss current research papers in Computer vision, Natural language processing, Reinforcement learning and Robotics.

Introduction to Spectral Graph Theory

Propose a video recognition framework that relies on two key observations: 1) While pixels may change to frame, the semantic content of …

Training network not in native parameter space, but instead in a smaller, randomly oriented subspace.

A novel meta-learniing algorithm that learns to assign weights to training examples based on their gradient directions.

Overview of Semantic Segmentation Architectures.

Exploiting the robustness of Residual Networks (ResNets) to layer dropping, The framework selects on-the-fly which residual blocks to …

Dense Convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion.

We further propose a novel training procedure which starts out with a dense convolution but gradually evolves towards a grouped …

Contact

  • Kohli Research Block, IIIT Hyderabad, Telangana, 500032, INDIA