Sonaal Kant

I am currently pursuing a Master's degree in Computer Science at the University of Maryland, College Park, where I am advised by Abhinav Shrivastava. My core research interests lie in the field of generative AI, and I am passionate about working on projects that can reach a large audience.

From 2017 to 2021, I worked as a Senior Data Scientist at ParallelDots, where I gained hands-on experience in Computer Vision and Natural Language Processing. I also had the opportunity to work on scaling and deployment of large models. In addition, I worked as a Research Associate under Dr. Anubha Gupta at IIIT, Delhi from 2018 to 2019. During this time, I explored the applications of computer vision in medical imaging diagnostics.

I completed my undergraduate degree in Information Technology at Maharaja Agrasen Institute of Technology in New Delhi, where I had the privilege of being advised by Dr. M.L Sharma for my Bachelor Thesis Project. During my undergraduate studies, I gained valuable experience working as a freelance consultant, helping and guiding other students with their projects.

CV  /  Email  /  GitHub  /  Google Scholar  /  LinkedIn

profile photo

Recent News

  • [April 23] Organising Visual Perception and Learning in an Open World Workshop at CVPR 2023
  • [May-Aug 22] Applied Scientist Intern at Amazon : Researched on estimating body measurements from monocular RGB image.
  • [Aug 21] Joined the MS CS program at UMD

Research

I'm interested in computer vision, machine learning, optimization, graphics and robotics.

Learning gaussian maps for dense object detection
Sonaal Kant
British Machine Vision Conference, 2020

Learning to predict small similar looking objects placed in close proximity by predicting gaussian map in a multi-task framework
LeukoNet: DCT-based CNN architecture for the classification of normal versus Leukemic blasts in B-ALL Cancer
Simmi Mourya*, Sonaal Kant*, Pulkit Kumar*, Ritu Gupta , Anubha Gupta
In submission

A deep learning framework for classifying immature leukemic blasts and normal cells by fusing Discrete Cosine Transform (DCT) domain features extracted via CNN with the Optical Density (OD) space features.
Benchmark for generic product detection: a low data baseline for dense object detection
Sonaal Kant*, Muktabh Mayank Srivastava*
Image Analysis and Recognition: 17th International Conference, ICIAR 2020

Benchmarking object detection methods in a low data regime and showing how augmentation based training can give similar performance in comparison with recent works.
Towards automated tuberculosis detection using deep learning
Sonaal Kant*, Muktabh Mayank Srivastava*
IEEE Symposium Series on Computational Intelligence (SSCI), 2018