Welcome! Glad you found me.

About me

I am a final-year Ph.D. student in the Computer science and engineering department at Michigan State University. My research interests lie in the field of Computer Vision, with particular interests in content authenticity, manipulation detection/localization, model parsing, generative modeling, and image generation. I also have some experience with object detection in the wild. All these approaches are based on a proactive scheme which involves adding a signal to the training images, followed by performing various downstream tasks. I’m working under the guidance of Dr. Xiaoming Liu at Computer Vision Lab (CVLAB).
I received my Bachelor’s degree in Electrical and Instrumentation Engineering from Birla Institute of Technology and Science, Pilani, India, in 2019. In summer 2023, I was a research scientist intern at Adobe, where I, along with Dr. John Collomosse and Dr. Shruti Agarwal worked on developing a proactive-based scheme for synthetic media provenance identifying which training images are most responsible for a Generative AI image.
I would be returning to Adobe in summer 2024 as a research scientist intern, focusing on media provenance! Delving into the realm of digital trust and authenticity, I’ll be working on project aimed at ensuring the integrity and credibility of digital media under Dr. John Collomosse and Dr. Shruti Agarwal.

Full-time opportunities

I’m looking for full time opportunities, preferrably in the areas of Responsible AI, media provenance, deepfake detection, etc. If you want to discuss my research or any collaboration idea, I’m just an email away :).

News

  • [2024/03/02]   ProMark, a proactive scheme-based causal attirbution method for content generated by GenAI models is accepted to CVPR 2024. This work is in collaboration with Adobe during my internship in summer 2023. Link
  • [2024/01/01]   Our survey paper based on reverse engineering of deceptions is published at Foundations and Trends in Privacy and Security. Link
  • [2023/09/22]   PrObeD, a proactive scheme-based wrapper to improve the performance of 2D object detection for generic and camouflaged objects got accepted to Neurips 2023. Link
  • [2023/07/29]   Our paper on reverse engineering of generative models (Model Parsing) is accepted to Transactions on Pattern Analysis and Machine Intelligence. Link
  • [2023/22/05]   Started working as a Research Scientist intern for Adobe during the summer 2023.
  • [2023/02/03]   Our paper on Proactive Manipulation Localization is accepted at CVPR 2023. Link
  • [2023/26/01]   Tech talk given in-person at Scale AI on Reverse Engineering of Generative Models. Link
  • [2022/31/08]   Virtual Tech talk given at Scale AI on Proactive Detection. Link
  • [2022/05/03]   Our paper on Proactive Manipulation Detection is accepted at CVPR 2022. Link
  • [2021/06/15]   In collaboration with Meta AI, our deepfake model parsing work is widely reported in CNBC, CNET, Engadget, Fortune, The Mac Observer, MSU Today, New Scientist, SiliconAngle, VentureBeat, The Verge, and Wall Street Journal.

Collaborators

I have collaborated with many experts in the computer vision field, namely Dr. Xiaoming Liu, Dr. Tal Hassner, Dr. John Collomosse, Dr. Shruti Agarwal, Dr. Xi Yin and Dr. Sijia Liu. I’m thankful for all the suggestions provided by everyone mentioned above.

Interests other than sitting in front of a laptop

I love going for long drives.
When having small breaks from work, I like solving jigsaw puzzles or reading a book.
Football, badminton, and table tennis are some of the outdoor sports I like.