Intern Spotlight: Lucy Cai uses machine learning to turn 2D images into 3D
What did you work on during your internship?
I worked with my group's Multimodal 3D Vision task, which is looking to convert between 2D image modalities such as electro-optical, infrared, and synthetic aperture radar, and 3D imagery such as that taken by a lidar device. My project involved using machine learning to convert 2D images into 3D point clouds. The neural network code I worked with takes in an RGB image and outputs a depth map and a point cloud estimation. I tested it on various datasets and analyzed its performance, as well as thought of methods to improve results. Though not part of my main project, I have also gotten the chance to work with hardware through helping collimate a laser.
Which aspect of your internship was the most helpful or interesting to you?
I loved learning about all of the different impacts each group's work in the Laboratory had. From disaster relief to combating drug trafficking, it was very cool to witness the extent of the power that these technologies could have. I appreciated unique opportunities such as touring the Flight Test Facility on the Air Force base and attending R&D seminars. Also, I loved getting to know the other interns. Though I didn't come in with any expectations, becoming friends with some of the most fun and passionate people I now know was one of the most rewarding parts of my summer.
What impact do you think this experience will have on your future?
I think Lincoln Lab incorporates the research-like environment of academia with the real-life impacts of industry, which is a rare combination. Now that I know it's possible to get the best of both worlds, I would like to pursue a career that will allow me to do similar work. I am also grateful to have gotten a taste of both software and hardware work, and my experiences on both sides will certainly guide me as I look towards finishing my degree in college.
To learn more about the Laboratory's student opportunities, click here.