About Me

I am a researcher in the Applied AI team as a part of the Advanced Tehnology Group at Dolby Laboratories in San Franscisco. Prior to this I graduated with my PhD from the DICE Lab at New York University under the supervision of Dr. Chinmay Hegde, in 2022. I have also interned at Google in 2021, Adobe in 2020 and MERL in 2018. I am an alumnus of Iowa State University and BITS Pilani, Goa, where I earned degrees in Electrical Engineering and Physics.

My interests lie in the intersection of signal processing, computer vision and deep learning. At Dolby I have primarily worked at enhancing video applications through deep learning approaches.

During the course of my PhD, I worked on several problems in inverse imaging, such as compressed sensing, phase retrieval and HDR imaging using deep neural networks. On the computer vision side, I have worked on developing algorithms to train adversarially robust neural networks. Concurrently I have also looked at new attack models to inspect vulnerabilities of vision based transformer architectures.

Outside of the topics that I explored during my PhD, I am also broadly interested in multi-modal learning, generative networks and self-supervised learning.

Updates:

April 2023: Our paper titled Can Deepfakes be created on a whim? was presented at the ACM 2023 Web Conference CySoc Workshop.

April 2023: Our paper titled A Few Adversarial Tokens Can Break Vision Transformers has been accepted to the CVPR 2023 Workshop on Adversarial Machine Learning on Computer Vision.

May 2022: Our paper on Token Attacks on Vision Transformers has been selected for spotlight presentation at CVPR 2022 Workshop on Transformers for Vision.

May 2022: Extended version of our paper on Inverse Imaging using Generative Priors has been accepted to IEEE Transactions on Information Theory.

March 2022: I will be giving a talk at the SIAM conference on Imaging Science on March 23.

February 2022: Our paper on Inverse Imaging using Generative Priors has been accepted to ICASSP 2022!