Experience

Computer Vision Research Engineer, LENS Corporation

Corrected and evaluated programs for fingerprint matching algorithms in C++ using metrics like ROC Curve, similarity scores, etc.
Developed a C++ based pipeline including OpenCV and Torchscript to deploy the DL model for fingerprint feature extraction on GPU and CPU-based systems.

Advised by Dr. Saket Anand and Dr. Sanjit Kaul.
Developed pipeline for cross-calibration between multiple cameras and LiDAR sensors.
Worked on developing and benchmarking Lane Detection models along with their deployment for autonomous cars.
Created ROS nodes for deploying lane detectors on a NVIDIA AGX Xavier as part of the autonomous stack of the car using C++, Torchscript and OpenCV.

Worked under the mentorship of Dr. Akshay Nambi.
Created solutions for various edge cases for Automatic Driver License Testing
Managed complete deployment of the project at IDTR Aurangabad.
Contributed to code refractoring for the deployment of the project at various places with the least possible effort.

Managed features and statistics design using matplotlib, plotly and seaborn. Developed interactive visualizations such as Heatmaps, and Geoplots based on the datasets recipes for the website. Worked on generation of a number of statistical figures for the recipe datasets.
Worked on the curation and compilation of databases through various sources and developed algorithms for the integration of databases for proper usage.

Worked under the mentorship of Dr. Manoj Sharma for Deep Learning based training and did research work for Image and Video Super-Resolution.
Contributed to building Deep Learning architectures including Deep Back Projection networks for Video Super-Resolution for NTIRE (New Trends in Image Restoration and Enhancement) Challenge 2019 in conjunction with CVPR.
The team secured 11th rank for Video Super-Resolution Challenge and got a Team Paper published with contribution of all the top rankers in the Challenge.

Google Sites

Report abuse