Burhan Ahmad Mudassar

Burhan Ahmad Mudassar

Ph.D Student

Georgia Institute of Technology

Biography

Burhan Ahmad Mudassar is a Ph.D candidate in the Gigascale Reliable and Energy-Efficient NanoSystems (GREEN) Lab at Georgia Institute of Technology. His research interests include developing novel paradigms for sensor control in smart cameras to achieve reliable task performance while conserving performance metrics such as bandwidth, power and energy. In addition, his research interests include developing deep learning based solutions to niche computer vision problems such as small object detection and moving camera action detection.

His spare time is spent traveling (a shared passion with his wife), creating imaginary worlds for his two kids and mindlessly scrolling the daily news feed. Love all kinds of fiction especially fantasy, magic and whodunits but he wishes he had more time to read.

Interests

  • Embedded Vision
  • Smart Cameras

Education

  • PhD in Electrical and Computer Engineering, Current

    Georgia Institute of Technology

  • Masters in Electrical and Computer Engineering, 2013

    Georgia Institute of Technology

  • BS in Electronics Engineering, 2012

    National University of Sciences and Technology (NUST) Pakistan

Skills

Embedded Development

2 years experience with RTL development on FPGA and Embedded DSP

Deep Learning

2 years experience with Pytorch/Tensorflow and 1 year experience with Caffe. Training and inference of computer vision models for object detection and action detection

Programming

4 years experience with Python, 3 years experience with C/C++ including OOP concepts, 2 years experience with Verilog for FPGA programming

Experience

 
 
 
 
 

Intern

Samsung Semiconductor Inc.

Jan 2020 – May 2020 California
Responsibilities include: * ISP and DNN pipeline optimization* Quantization of Pytorch Models
 
 
 
 
 

Graduate Research Assistant

Georgia Institute of Technology

Jan 2017 – Present Georgia
GREEN Lab
 
 
 
 
 

Sr. Design Engineer

Center for Advanced Research in Engineering (CARE Pvt Ltd)

May 2015 – Aug 2016 Islamabad, Pakistan
As part of the Software Defined Radio Group, my responsibilities include system development, debugging and implementation of proprietary waveforms on multiple platforms including DSPs and FPGAs. Some of the tasks that I have completed are: * Hardware/Software Co-Design of waveform transceivers on FPGA+DSP * Implemented CPM Coarse Timing Recovery, Carrier Phase and Frequency Offset Module in FPGA * Implemented GMSK Frequency Hopping Modulator on TI TMS320C64X DSP

Accomplish­ments

Gold Medal for Best Research Project in Undergraduate Studies

National University of Sciences and Technology Jan 2014
Gold Medal for Best Research Project in Undergraduate Studies - Multi-touch Interactive Table-Top

Silver Medal for Second Highest GPA in Undergraduate Studies

National University of Sciences and Technology Jan 2014
Silver Medal for Second Highest GPA in Undergraduate Studies

Fulbright Scholarship for Graduate Study

US Educational Foundation in Pakistan (USEFP) Aug 2013
Fulbright Scholarship for Masters Study in Electrical and Computer Engineering at Georgia Tech

Recent Posts

Power of Hooks in Pytorch

What are hooks? Pytorch allows you to add custom function calls to its module and tensor objects called hooks. The calls can both be added to the forward method of the object as well as the backward method.
Last updated on Jul 22, 2020 4 min read

Projects

Small Object Detection

In this work, we present a simple solution for increasing small object detection performance. Through a series of empirical experiments we analyze the effect of aggressive down-scaling of feature maps in a convolutional backbone.

Recent Publications

Quickly discover relevant content by filtering publications.

WarningNet: A Deep Learning Platform for Early Warning of Task Failures under Input Perturbation for Reliable Autonomous Platforms

Minah Lee, Burhan A Mudassar, Taesik Na, Saibal Mukhopadhyay

MagNet: Discovering Multi-agent Interaction Dynamics using Neural Network

Priyabrata Saha, Arslan Ali, Burhan Mudassar, Yun Long, Saibal Mukhopadhyay

Rethinking Convolutional Feature Extraction for Small Object Detection

Burhan A Mudassar, Saibal Mukhopadhyay

A Camera with Brain - Embedding Machine Learning in 3D Sensors

Burhan A Mudassar, Priyabrata Saha, Yun Long, Mohammed F Amir, Taesik Na, Jong Hwan Ko, Marilyn Wolf, Saibal Mukhopadhyay

A Spatiotemporal Pre-processing Network for Activity Recognition Under Rain

Minah Lee, Burhan A Mudassar, Taesik Na, Saibal Mukhopadhyay
See all publications

Popular Topics

Object Detection Deep Learning Dropout Pytorch Small Object Detection

Contact