hit counter
Viraj Shah
Viraj Shah

Viraj Shah

Research Scientist, Google

I am Research Scientist at Google in Computational Imaging team. Previously, I earned PhD from UIUC where I worked with Dr. Svetlana Lazebnik on Generative Models and Computer Vision.
RESUME GOOGLE SCHOLAR EMAIL PROJECTS

I am open to collaboration opportunities broadly in the area of generative models/diffusion models/image editing. Our team at Google may also have internship openings for Ph.D. students time to time. Feel free to write me an email if you're interested!

I am Research Scientist at Google in Computational Imaging team. I obtained my PhD in Electrical Engineering at University of Illinois, Urbana-Champaign where I worked with Dr. Svetlana Lazebnik on Generative Models.

I mainly work on leverage novel Generative Models such as Diffusion Models and GANs in solving image manipulation and imaging inverse problems. Please check out the Projects and Publications section, or my resume to learn more!

Previously, I completed B.Tech. in EE from IIT Roorkee in 2016 and joined PhD program at Iowa State University where I earned MS in ECE before transferring to UIUC. At ISU, I worked at DICE Lab with Dr. Chinmay Hegde.

Throughout the course of my studies, I have had the privilege of working with and receiving guidance from amazing researchers and teams:

Throughout the course of my studies, I have had the privilege of working with and receiving guidance from amazing researchers and teams:

Projects & Publications


Generative Models

project image
New!

UnZipLoRA: Separating Content and Style from a Single Image

Chang Liu, Viraj Shah, Aiyu Cui, Svetlana Lazebnik
paper | project page

project image
Accepted to ECCV 2024!

ZipLoRA: Any Subject in Any Style by Effectively Merging LoRAs

Viraj Shah, Nataniel Ruiz, Forrester Cole, Erika Lu, Svetlana Lazebnik, Yuanzhen Li, Varun Jampani
ECCV, 2024
paper | project page | poster

project image
Best Paper Award at CVFAD Workshop, CVPR 2024!

Street TryOn: Learning In-the-Wild Virtual Try-On from Unpaired Images

Aiyu Cui, Jay Mahajan, Viraj Shah, Preeti Gomathinayagam, Svetlana Lazebnik
WACV 2025
CVFAD Workshop, CVPR, 2024
paper | project page | video | poster

project image

MultiStyleGAN: Multiple One-shot Image Stylizations using a Single GAN

Viraj Shah, Ayush Sarkar, Sudharsan Krishna, Svetlana Lazebnik
paper | project page | code

project image

Generative AI: Challenges and Opportunities in the Context of India

Viraj Shah, Kartik Patel
EC3V Workshop, CVPR, 2023
paper | poster

project image

Make It So: Steering StyleGAN for Any Image Inversion and Editing

Anand Bhattad, Viraj Shah, Derek Hoiem, David A. Forsyth
paper | project page

project image

Near Perfect GAN Inversion

Qianli Feng, Viraj Shah, Raghudeep Gadde, Pietro Perona, Aleix Martinez
paper | slides

project image

Encoding Invariances in Deep Generative Models

Viraj Shah*, Ameya Joshi*, Sambuddha Ghosal, Balaji Pokuri, Soumik Sarkar, Baskar Ganapathysubramanian, Chinmay Hegde
paper

project image

InvNet: Encoding Geometric and Statistical Invariances in Deep Generative Models

Ameya Joshi, Minsu Cho, Viraj Shah, Balaji Pokuri, Soumik Sarkar, Baskar Ganapathysubramanian, Chinmay Hegde
AAAI, 2019
paper | code

project image

Generative Models for Solving Nonlinear Partial Differential Equations

Ameya Joshi*, Viraj Shah*, Sambuddha Ghosal, Balaji Pokuri, Soumik Sarkar, Baskar Ganapathysubramanian, Chinmay Hegde
MLPS Workshop, NeurIPS, 2019
paper

project image
Second Prize at MRS Open Data Challenge 2019!

Physics-aware Deep Generative Models for Creating Synthetic Microstructures

Rahul Singh*, Viraj Shah*, Balaji Pokuri, Soumik Sarkar, Baskar Ganapathysubramanian, Chinmay Hegde
MMLS Workshop, NeurIPS, 2018
paper | poster | slides | news


Imaging Inverse Problems

project image

Sparse Signal Recovery from Modulo Observations

Viraj Shah, Chinmay Hegde
EURASIP Journal on Advances in Signal Processing, 2020
paper | code | slides

project image

Signal Reconstruction from Modulo Observations

Viraj Shah, Chinmay Hegde
GlobalSIP, 2019
paper | code | slides

project image

Alternating Phase Projected Gradient Descent with Generative Priors for Solving Compressive Phase Retrieval

Rakib Hyder, Viraj Shah, Chinmay Hegde, Salman Asif
ICASSP, 2019
paper | code | poster

Extends the projected gradient descent-based approach to leverage Generative Priors for phase retrieval.

project image

Solving Linear Inverse Problems using GAN Priors: an Algorithm with Provable Guarantees

Viraj Shah, Chinmay Hegde
ICASSP, 2018
paper | code | poster | slides

Proposes a provable projected gradient descent-based approach to leverage Generative Priors for compressive sensing.

project image

Reconstruction from Periodic Non-linearities, with Applications to HDR Imaging

Viraj Shah, Mohammadreza Soltani, Chinmay Hegde
Asilomar, 2017
paper | poster

Reconstructing HDR image from multiple modulo images in compressive sensing setting.


Miscellaneous

project image

CloudFindr: A Deep Learning Cloud Artifact Masker for Satellite DEM Data

Kalina Borkiewicz, Viraj Shah, JP Naiman, Chuanyue Shen, Stuart Levy, Jeff Carpenter
VIS, 2021
paper

project image

Quantifying driver speed behavior from real-time physiology in type 1 diabetes

Viraj Shah, Jennifer Merickel, Pranamesh Charkraborty, Anuj Sharma, Chinmay Hegde,..., Matt Rizzo
Fast-ZERO, 2019
paper