Jaydeep Chauhan
Development
North Carolina, United States
Skills
Machine Learning (ML)
About
Jaydeep Chauhan's skills align with IT R&D Professionals (Information and Communication Technology). Jaydeep also has skills associated with Programmers (Information and Communication Technology). Jaydeep Chauhan appears to be an entry-level candidate, with 16 months of experience.
Work Experience
Associate Data Scientist
ParallelDots, Gurugram
November 2020 - July 2021
- Object Detection Benchmarking: Led the benchmarking of Domain Invariant Object Detectors across diverse client datasets, contributing to the enhancement of the ShelfWatch product. Evaluated and implemented state-of-the-art object detection models to replace the generalized object detector. ◦ Managed and optimized Computer Vision pipelines:: Effectively oversaw and enhanced end to end Computer Vision pipelines for leading FMCG companies, including BAT, RB and Mondelez. Led client communications, addressing their requirements, and continuously improved pipelines to align with client expectations. ◦ Object Detection for Mobile Device: Trained and optimized object detection models for mobile devices. ◦ Research Work: Made significant contributions to ongoing research in semi-supervised learning for dense object detection, resulting in publication at the CVPR 2021 RetailVision Workshop.
Computer Vision Research Intern
Vimaan Robotics, Santa Clara
June 2022 - September 2022
- Image Quality Assessment (IQA): Developed a cutting-edge 'No Reference Image Quality Assessment' method utilizing deep learning exclusively trained on synthetic data. This innovation accurately assesses image quality and pinpoints distorted regions in multi-task settings, and exhibits robust generalization capabilities when applied to real-world data. ◦ Data valuation: Worked on data valuation methods by implementing a meta-reinforcement learning approach, quantifying sample quality. This significantly contributed to enhancing the robustness of the data utilized in various projects. ◦ Image enhancement: Played a key role in developing a pipeline for image enhancement using state-of-the-art deep learning methods for image deblurring.
Software Engineering Intern
Bhaskaracharya Institute For Space Applications and Geo-Informatics(BISAG)
December 2018 - March 2019
- Project: Developed a decentralized chatting application for BISAG scientists using the open-source framework 'Matrix Synapse, ' showcasing proficiency in Python, Angular, and SQLite. Projects • Volumetric rendering using Neural Radiance Fields: Implementation of NeRF(https://www.matthewtancik.com/nerf) to render a 3D view of a scene • Supervised Contrastive Learning for pretrained language models: Implementation of supervised contrastive learning for pretrained language models paper and analyzed the embeddings(https://arxiv.org/abs/2011.01403). • Speech Enhancer: This project is about enhance a audio file using end to end deep learning system. SEGAN(Speech Enhancement Generative Adversarial Network) was used (Aug '20). • Autoregressive Generative Models: Implementation of Autoregressive Generative Models such as PixelCNN and PixelCNN++ on Standford Dog Dataset (August '20). • Matrix optimization using Angular and Python: Developed a decentralized chat application based on popular open-source matrix protocol and synapse server using Angular as front-end and Python-twisted as a backend (March '19). • Duplicate Image Detection Tool: A PyQt based GUI application that can detect Duplicate Images, able to detect noise in Images and delete them (Feb '18). • Music Recommendation Engine: A music hosting site and music recommendation engine built in C# using collaborative
Tech
October 2017 - October 2017
- filtering (Oct '17) • Web crawler and Information retrieval system: Small project which crawls websites and store information in database. User can retrieve information using search engine and system shows relevant web pages according to the query. Tech: Php,
Ajax
February 2017 - February 2017
Education