I have been associated with the Robotics Research Center for almost two years now, working on projects with Prof. Madhava Krishna that involve a wondrous blend of vision and robotics. Most recently, our work titled ‘Deep Model Predictive Control for Visual Servoing’ was accepted at the Conference on Robot Learning, 2020! We proposed a novel image-based visual servoing framework which processes visual features, trains on-the-fly and generates optimal control to successfully servo the environment.

I work closely with Dr. Harit Pandya at Toshiba Research Europe, Cambridge. I also collaborate with Dr. Brojeshwar Bhowmick and Gourav Kumar at the TCS Innovation Labs, Kolkata. Currently, I am working with Prof. Avinash Sharma on human motion synthesis and neuromorphic vision. We are working on the generation of asynchronous event trajectories using photometric feature tracking in order to capture high-speed motions from event cameras.

Quadcopters and drones at the Robotics Research Center, IIIT Hyderabad.
Beautiful campus of IIIT Hyderabad! Really miss this place!


Past Experience

During the Summer of 2020, I worked at Siemens in the corporate research division. I was associated with the visual perception lab, where I worked heavily in pose estimation and its extension to action recognition.

In Spring 2020, I worked on a research project in face detection advised by Prof. PJ Narayanan at CVIT. My work involved blending machine learning models with computer vision for society related applications. I also worked on a project with Prof. Avinash in 3D semantic scene segmentation, which involved combining 2D feature maps with 3D voxel data.

In Fall 2018, I worked as a research intern at the Virtual Labs, a social initiative of the government of India. I worked with my friend Gaurav Batra and developed full fledged experiments and interactive artefacts for various data structures and algorithms using software engineering research principles.

I am deeply passionate about computer vision, with a particular fondness for 3D scene understanding and object recognition. As part of the vision and AI courses at my university, I’ve had the opportunity to work on some very engaging projects. Do check out my GitHub for some interesting ones!