Navid Panchi

Explore My Work

I'm Masters student at Friedrich Alexander University Erlangen-Nuremberg. My Research interest lies at the intersection of Computational Engineering, Materials Science, High Performance Algorithms and Deep Learning.



  • Deep Learning-Based Stair Segmentation and Behavioral Cloning for Autonomous Stair Climbing, International Journal of Semantic Computing Vol. 13, No. 4 (2019) 1–16, World Scientific Publishing Company,  DOI:10.1142/S1793351X19003526, link.

  • Estimate Sequential Poses for Wireless Endoscopic Capsule Based on Encoder-Decoder Convolutional Neural Network towards Autonomous Navigation, Data Analytics in Biomedical Engineering and Healthcare(ELSEVIER), link.

  • Data Efficient Stagewise Knowledge Distillation, arXiv:1911.06786, link.



May 2019 - July 2019

L&T Technology Services: Research Internship

I worked on designing a multi-dimensional (2D + 1D) Convolutional Neural Network for extraction of Signature from machine audio. These Signatures were used for monitoring the state of the bearings as well as for remaining life estimation.

Dec 2018 - Jan 2019

GULUKUL, CA, USA: Deep Learning Intern

I worked on extending the paper "Everybody Dance Now" for content generation. Pix2Pix GAN was reimplemented and trained on OpenPose human keypoints for generation of cartoon characters to be used in children's comics.

May 2018 - July 2018

NUS Singapore: Research Internship

Worked towards design and implementation of a Convolutional Neural Network for prediction of 6 DOF pose of an endoscopic capsule. This work was converted into a research paper and submitted to ROBIO - 2019



July 2019 - Ongoing

B. Tech Final Year Project (Thesis) : Prediction of Thermodynamic Parameters of Different BMG Compositions using Machine Learning

We are working towards prediction of Glass Transition Temperature, Onset Crystallization Temperature and Liquidus Temperature of different bulk metallic glass forming compositions using machine learning.

Dec 2017 - Jan 2018

Behavioural Cloning for Prediction of Steering Angles (using Udacity's Simulator)

This work involves training of a simple 9 layered Convolutional Neural Network for prediction of steering angles of a car given an RGB image from the front camera. This is implemented in TF-Learn wrapper on Tensorflow Library.


Get in Touch

This site was designed with the
website builder. Create your website today.
Start Now