Rahul Dey
Hi! I am a Machine Learning / Computer Vision researcher and I completed my PhD in 2023 from the Department of Computer Science, Michigan State Univeristy. I was advised by Prof. Vishnu Boddeti in the Human Analysis Lab, with Prof. Xiaoming Liu, Prof. Arun Ross and Dr. Felix Juefei-Xu serving in my dissertation committee. I did my undergraduate and masters in Electrical and Electronic Communication Engineering from the Indian Institute of Technology (IIT), Kharagpur between 2008 - 2013.
My research is centered around Generative AI and 3D Computer Vision. My PhD thesis is on 3D Face Modeling and its Applications in Generative Tasks and Occlusion-aware Reconstruction. I am a passionate learner and keen observer of details in research problems with strong interpersonal and presentation skills.
I spent the summer of 2022 in the Computer Vision group at Mitsubishi Electric Research Lab (MERL), Cambridge, MA, working with Dr. Tim Marks, Dr. Ye Wang, and Prof. Bernhard Egger on a disentangled and controllable face neural implicit 3D-GAN.
During 2015-2016, I worked as a Research & Development Engineer (Image Processing) at a start-up Logic Fruit Technologies, India where I designed a Real-time Object Tracking and Panorama Generation algorithm in an embedded device for DRDO, India. Before that, during 2013-2015, I worked as a Software Engineer at Electronic Arts where I was credited for the game titled Need for Speed 2016.
Email / Resume / LinkedIn / GitHub / Google Scholar
Undergoing Projects
1. CoLa-SDF: Controllable Latent StyleSDF for Disentangled 3D Face Generation
Neural implicit 3D-GAN of faces with multi-view consistency and disentangled latent space for shape, albedo, illumination, hairstyle allowing explicit control and editability over these factors
US Patent Filed
2. Diffusion3DFace: Occlusion-aware Diverse Textured 3D Face Reconstruction with Diffusion
Diverse texture and 3D shape reconstruction given a single occluded face image, using CLIP-based latent diffusion, a Mesh GNN and a new diversity-oriented DDPM sampler.
Publications
Rahul Dey, Felix Juefei-Xu, Vishnu Boddeti and Marios Savvides, ACCV 2018
*Oral, Best Student Paper Award
pdf / supplementary / project / arxiv / code
Gate-Variants of Gated Recurrent Unit (GRU) Neural Networks
Rahul Dey and Fatih Salem, MWSCAS 2017
Experiences
May 2022 - March 2023
2015 - 2016
2013 - 2015
Professional Activities and Teaching
Reviewer
Teaching Assistant
Introduction to Machine Learning, Prof. Parisa Kordjamshidi, Department of Computer Science and Engineering, Michigan State University, 2022
Introduction to Machine Learning, Prof. Kristen Johnson, Department of Computer Science and Engineering, Michigan State University, 2021
Introduction to Artificial Intelligence, Prof. Vishnu Boddeti, Department of Computer Science and Engineering, Michigan State University, 2020
Introduction to Machine Learning, Prof. Jiyau Zhou, Department of Computer Science and Engineering, Michigan State University, 2019