Stalk-Net
Why We're Great >
A generic deep-learning based pipeline for stalk width-estimation. Faster-RCNN based region proposals are used for semantic segmentation of stalks. Stereo Imagery is used to get metric width.
Stalk-Net
Why We're Great >
A generic deep-learning based pipeline for stalk width-estimation. Faster-RCNN based region proposals are used for semantic segmentation of stalks. Stereo Imagery is used to get metric width.
Unsupervised Depth Estimation with GANs
Why We're Great >
A generic deep-learning based pipeline for stalk width-estimation. Faster-RCNN based region proposals are used for semantic segmentation of stalks. Stereo Imagery is used to get metric width.
Hi! I'm Tanvir Parhar
I am a Research Associate at Carnegie Mellon University, actively working on agricultural automation and perception. I am also interested in vehicular autonomy. I believe there will be a day when cars will drive just using images (we humans do it!) and the path to that will be a combination of known control algorithms and current state of the art Deep Reinforcement Algorithms.

Tanvir Pal Singh Parhar
Associate Researcher II and Master's Student in Robotics at Carnegie Mellon University
Phone:
+1 412-999-9647
Email:
Office Address:
Room 1111, Newell Simon Hall,
Carnegie Mellon University
EXPERIENCE
Oct 2016 - Current
Associate Researcher II
CARNEGIE MELLON UNIVERSITY
As an Associate Researcher I work in agricultural robotics and perception. This ranges from developing deep-learning algorithms to detect fruits, pests and diseases and correlating them to yield. Using stereo vision to compute phenotypes like plant leaf area, stalk width, etc. I also work on optimizing the algorithms to run on small form-factor devices like Jetson-TX2
Jan 2016- Jun 2016
Intern
CARNEGIE MELLON UNIVERSITY
As an Intern my job was to develop algorithms to detect leaves and compute canopy cover in grape vines.
EDUCATION
2018-2020
Master's Degree
Master of Science in Robotics | Carnegie Mellon University
As a Master's student in Robotics my interests are in computer vision and Reinforcement learning. I am working in learning vision based controllers for autonomous driving. I am interested in developing neural-network controllers that trained via a combination of Reinforcement Learning and Optimal Control strategies.
2012-2016
Bachelor's Degree
Bachelor of Technology in Electronics and Instrumentation | VIT Vellore
As an undergraduate in Electronics and Instrumentation I worked on many different projects, ranging from developing inventory management robots to developing UAVs for government organizations. I lead the UAV Development Center of VIT University during my final year of college.
2010-2012
High School Diploma
St. Paul's Sr.Secondary High School
I Graduated high-school with a distinction by scoring highest in Mathematics and Computer Programming in my class.
SKILLS

ROS- Robot Operating System
Machine Learning - Tensorlfow/Caffe
Computer Vision- OpenCV, Matlab
Robot Autonomy
Micro-controller programming
On-board Deep NN deployment