EXPERIENCES

-- Worked on alternate architectures for Novel View Synthesis and Neural Rendering at Intel Labs. Worked on a GPU-Guided approach for Neural Rendering which was submitted to SIGGRAPH 2021. Proposed an efficient architecture for Video style transfer and filed a provisional patent with the USPTO on a novel application. Deployed the architecture onto ONNX_Runtime optimized with OpenVINO.

-- Worked on Barcode Decoding using Cameras for warehouse boxes. The entire project was structured as a new problem and the datasets, models and, augmentations were done from scratch. Also ported an object detection network on an embedded device - assisting autonomous navigation and frame reduction using TensorRT.

-- Domain Adaptation for Object Detection at NTU Singapore with Dr. Lam Siew Kei. Did a thorough literature survey of Domain Adaptation. Worked on a robust object detector based on Faster-RCNN that can perform well across different datasets.

-- Worked on Vision-based Remote Monitoring of Vital signs in the context of COVID-19. Different combinations of Learning based and signal processing methods were tried, resulting in a paper accepted at Springer Computing Journal.

-- At the end of my sophomore year, I have worked in Infinity Labs of UST Global which takes in research-grade projects and transforms them into scalable projects. The project on Simplifying the online loan process resulted in a publication.

-- Worked with Dr.Pratik Narang on the entire Computer Vision pipeline for Autonomous Vehicles. As a part of this, I have worked on Optical Flow, Semantic Segmentation using CNNs, and Single Image Deraining using GANs. This work has been submitted to Springer Computing Journal

-- Worked in ADAPT Lab under Prof. Navneet Goyal on Deep Learning solutions for Biodiversity Awareness in society. Specifically worked on Sketch-based Image Retrieval of images of plants and animals.

-- Have also done Interdisciplinary research (Deep Learning for Chemistry) with Graph Convolutional Networks for solving chemical problems.

PUBLICATIONS

-- AI-enabled remote monitoring of vital signs for COVID-19: Methods, Prospects and Challenges, Springer Computing Journal (accepted)

-- Complete Scene Parsing for Autonomous Navigation in Unstructured Environments, 3rd IEEE International Conference on Intelligent Autonomous Systems, NTU, Singapore, 2020 (accepted).

-- Social Network Analysis using Data Segmentation and Neural Networks International Research Journal of Engineering and Technology (IRJET) Volume 5, 2018 (accepted).

-- DerainGAN: Single Image Deraining Using Wasserstein GAN, Springer Multimedia Tools and Applications Journal (in review)

TECHNICAL TEAM

-- Founding member of a student Self Driving Car Research team for highly complex Indian roads (in which I am currently involved as a Senior Computer Vision Researcher). We are working on making a standard gasoline car (Maruti Alto K-10) into an Autonomous Car.

-- We have achieved Level II autonomy(implementing lane-keeping and adaptive cruise control) and are continuously publishing papers. As a part of this, I have published a paper at IEEE ICoIAS on Scene Parsing in Unstructured Environments

RELEVANT COURSES

Neural Networks and Fuzzy Logic(Audited and Teaching Assistant), Data Mining, Data Structures and Algorithms, ObjectOriented Programming, Database Systems, Computer Networking, Operating Systems, Compiler Design, Probability and Statistics, Linear Algebra(with Matrix theory), Deep Learning Specialization(Online), CS231n(Online)

HACKATHONS

-- (Yes Bank Datathon) Led the team to develop an ensemble ML model for solving the Address Geocoding problem. Stood 1st runner up, winning a prize of INR 30000

-- (Flipkart GRiD) Worked on Dual Path Networks for Object Localization(Level 3) achieving 88% accuracy on Flipkart custom product dataset

August 2020 Attended Google Research India AI Summer School (Computer Vision Track)