I am particularly interested in Computer Vision and Machine Learning research focusing on the following directions:

• Vision and Language: Can we build efficient end-to-end and multi-task visio-linguistic models capable of capturing correlations between regions in images, actions in videos with corresponding phrases in sentences?

• Multi-modal Representation Learning: How can we best build models which leverage features from several modalities allowing autonomous agents to navigate and perceive the environment around them in a more holistic fashion.

• Unsupervised Robotic Perception: Considering the cost and lack of good quality labelled data, how can we learn from easily available unlabelled visual data by using self-supervised, semi-supervised, or few-shot learning techniques ?

I have been lucky to be supervised by Prof. Partha Pratim Das for my B.Tech thesis on Detecting and Tracking Feet Movement in Indian Classical Dances and by Prof. Animesh Mukherjee for my M.Tech thesis on Multimodal Hate Speech Detection in tweets.

I have also had the chance to work under the esteemed guidance of Prof. Ram Nevatia and Prof. Florian Shkurti during academic research internships at University of Southern California and University of Toronto respectively.