UROP Openings

Have a UROP opening you would like to submit?

Please fill out the form.

Submit your UROP opening

Seeing Around Corners: A real life application


Term:

IAP

Department:

MAS: Media Arts and Sciences

Faculty Supervisor:

Ramesh Raskar

Faculty email:

raskar@mit.edu

Apply by:

Jan 10 2021

Contact:

Subhash Sadhu: subhashs@mit.edu

Project Description

The project is to enable a real life application to see around corners. The student would be able to employ reconstruction algorithms on cutting edge LIDAR and acoustic ranging systems to reconstruct the shapes of hidden objects. Our lab is renowned for inventing this technology. A TED talk by our advisor is here for context: https://youtu.be/Y_9vd4HWlVA . This technology has come a long way ever since and the student has a good opportunity to not only learn the whole system by implementing it in novel ways, they get to be a part of the process in pushing the boundaries. We aim to create a new age health imaging modality. Actual work will be a subset of: - Setting up and calibrating sensory LIDAR/SONAR system - Coding for data acquisition and interface - Implementing reconstruction algorithms involving high-end signal processing and machine learning We aim to publish a paper based on the student's work.

Pre-requisites

- Python: Numerical computing with Numpy and Scipy, Pytorch - Basics of Machine learning: Autodifferentiation, Neural Networks - Strong understanding of signal processing - Strong understanding of the physics of wave propagation for light, sound, waveguides. Must have done a course in optics or related topic to gain sufficient depth in this area. - Optimization methods