Research & education
Learning from images, language, and structure. Research threads from Carnegie Mellon, IIT Madras, IBM Research, and video classification.
Learning depth from video
As a research intern in the Machine Learning Department in 2017, I worked on geometry-aware deep learning for unsupervised depth and motion estimation from monocular video. S.N. Bose Scholar.
Finding structure in unstructured text
Worked on unsupervised text mining, with research on discovering events and entities in unstructured text.
Dialogue across images and words
My thesis studied multimodal visual dialogue. Other research explored word representations and polysemy in word2vec.
Recognising what happens in video
Built an ensemble of deep networks for multi-label video classification in the YouTube-8M Challenge.
A foundation in computer science
B.Tech + M.Tech in Computer Science and Engineering, with a focus on machine learning. Branch change to Computer Science. Ranked first in the freshman cohort. Star Teaching Assistant Award.