Scalable Learning for Object Detection with GPU Hardware

We consider the problem of robotic object detec- tion of such objects as mugs, cups, and staplers in indoor envi- ronments. While object detection has made significant progress in recent years, many current approaches involve extremely complex algorithms, and are prohibitively slow when applied to large scale robotic settings. In this paper, we describe an object detection system that is designed to scale gracefully to large data sets and leverages upward trends in computational power (as exemplified by Graphics Processing Unit (GPU) technology) and memory. We show that our GPU-based detector is up to 90 times faster than a well-optimized software version and can be easily trained on millions of examples. Using inexpensive off-the-shelf hardware, it can recognize multiple object types reliably in just a few seconds per frame. Authors: Adam Coates, Paul Baumstarck, Quoc Le, Andrew Y. Ng (2009)
AUTHORED BY
Adam Coates
Paul Baumstarck
Quoc Le
Andrew Y. Ng

Abstract

We consider the problem of robotic object detec- tion of such objects as mugs, cups, and staplers in indoor envi- ronments. While object detection has made significant progress in recent years, many current approaches involve extremely complex algorithms, and are prohibitively slow when applied to large scale robotic settings. In this paper, we describe an object detection system that is designed to scale gracefully to large data sets and leverages upward trends in computational power (as exemplified by Graphics Processing Unit (GPU) technology) and memory. We show that our GPU-based detector is up to 90 times faster than a well-optimized software version and can be easily trained on millions of examples. Using inexpensive off-the-shelf hardware, it can recognize multiple object types reliably in just a few seconds per frame.

Download PDF

Related Projects

Leave a Reply

You must be logged in to post a comment