I’m a college graduate of 2023 with an Integrated MSc in Mathematics and Computing from the Indian Institute of Technology (IIT) Kharagpur. I’m interested in Computer Vision, Computer Graphics and Simulations.
I have always been fascinated by the ability of computers to capture and recreate reality. I am currently working on the 3D reconstruction of large-scale environments using aerial photography.
Preimage
Preimage is 3D-as-a-service company. It has developed a fully-managed cloud photogrammetry application, along with several other novel technologies in the field of 3D reconstruction. I have been working at Preimage as a Research Engineer for Computer Vision.
I have worked on and improved several structure-from-motion components including feature detection and matching, motion averaging, and bundle adjustment. Consequently, I am extremely familiar with the SfM pipeline along with the codebases of some of the open-source repositories available for it. I have also worked on several novel view synthesis and neural surface reconstruction algorithms.
Working in a start-up environment has given me the vision and decision-making necessary to get quality results in a competitive timeline.
Intel Labs
Intel Labs is a global research organization committed to discovering and developing new technologies and compute forms to unleash the exponential power of data. I was a Software Development Intern of the team that develops and manages the Open3D library. Open3D is an open-source library that supports rapid development of software that deals with 3D data.
At Open3D, I worked on several features and fixes including adding a command-line interface for the library, developing a benchmarking system against competitor libraries and other quality of life changes.
The Open3D team maintains a very high coding standard for their library, and as a result the code for the library is clean and well organized. These coding and design standards have greatly inspired me to write clean, readable and organized code.
AGV-IITKGP Research Group
Autonomous Ground Vehicle (AGV), led by Professor Debashish Chakraborty, is a multi-disciplinary research group aimed at building a fully operational self-driving car. I was a part of this research group as an Undergraduate Researcher, and I led the Simultaneous Localization and Mapping team.
At AGV, from working on an in-house manufactured robot for an international competition to running my code on an actual self-driving car, I have gained significant hands-on experience in robotics. I have mapped a major chunk of the IITKGP campus using the Velodyne HDL-32E LiDAR and benchmarked several LiDAR localization algorithms. My team and I have modified a popular LiDAR localization package (LeGO-LOAM) to add re-localization capability in a prerecorded map.
Working on various projects at AGV, I had great exposure to various industry standard sensors and hardware. The friendly, supportive, and research-intensive environment that I found at AGV has encouraged me to try and recreate such an environment wherever I work.