Alan De-Hao Cheng
MIT EECS | Analog Devices Undergraduate Research and Innovation Scholar
A Depth-Estimation Platform for Low Power Time-of-Flight Imaging
2018–2019
Electrical Engineering and Computer Science
- Graphics and Vision
Vivienne Sze
Depth sensors are useful in a variety of mobile applications, ranging from robotics to augmented reality, by providing 3D information about a scene. One type of depth sensor is a time-of-flight (TOF) camera, which measures depth by emitting light and measuring its round-trip time. TOF cameras are appealing because they obtain dense depth measurements with minimal latency. However, one drawback is its illumination source, which is power hungry and limits the battery life of mobile devices. The goal of this project is to reduce the power consumption of depth sensing using TOF cameras by obtaining depth in a more energy-efficient manner. We propose to build a system that uses RGB images from a digital camera and previous depth measurements to estimate depth without illuminating the scene.
Since high school, I’ve been interested in working in both FPGA prototyping and computer vision. This SuperUROP opportunity, found through my final project in 6.111 (Introductory Digital Systems Laboratory), perfectly combines both of these fields. I hope to further my knowledge not just in these two topics but also in their fusion how hardware can be used to efficiently help improve image processing.