Gregory P. Meyer

Gregory P. Meyer, Ph.D.

Senior Software Engineer
Uber Advanced Technologies Group
Email:

Formerly, a graduate student at the University of Illinois at Urbana-Champaign where I was advised by Minh N. Do. Currently, a software engineer on the perception team at Uber ATG. My research interests are computer vision and pattern recognition with 3D sensors, and parallel computation for real-time execution of vision algorithms.

Education

University of Illinois at Urbana-Champaign, Aug. 2014 - Dec. 2016: Doctor of Philosophy, Computer Engineering; GPA: 3.92/4.00

University of Illinois at Urbana-Champaign, Aug. 2011 - May 2014: Master of Science, Computer Engineering; GPA: 3.91/4.00

University of Illinois at Urbana-Champaign, Jan. 2008 - Dec. 2010: Bachelor of Science, Computer Engineering; GPA: 3.55/4.00

Projects

	Online Face Verification with a Consumer Depth Camera We developed a system for accurate real-time 3D face verification using a low-quality consumer depth camera. To verify the identity of a subject, we built a high-quality reference model offline by fitting a 3D morphable model to a sequence of low-quality depth images. At runtime, we compare the similarity between the reference model and a single depth image by aligning the model to the image and measuring differences between every point on the two facial surfaces. The model and the image will not match exactly due to sensor noise, occlusions, as well as changes in expression, hairstyle, and eye-wear; therefore, we leverage a data driven approach to determine whether or not the model and the image match. [Paper] [Code]
	Robust Model-based 3D Head Pose Estimation We developed a method for accurate head pose estimation using a commodity depth camera. We perform pose estimation by registering a morphable face model to the measured depth data, using a combination of particle swarm optimization (PSO) and the iterative closest point (ICP) algorithm. [Paper] [Code]
	Improving Face Detection with Depth Face detection serves an important role in many computer vision systems. Without prior information, the size and position of a face within the image is unknown; therefore, the detector must exhaustively search the image at every position and scale. We developed a technique that leverages depth information to identify regions within a color image that may contain a face, as well as, estimate the size of the face. [Paper] [Code]
	3D GrabCut: Interactive Foreground Extraction for Reconstructed 3D Scenes In the near future, mobile devices will be able to measure the 3D geometry of an environment using integrated depth sensing technology. This technology will enable anyone to reconstruct a 3D model of their surroundings. Similar to natural 2D images, a 3D model of a natural scene will occasionally contain a desired foreground object and an unwanted background region. Inspired by GrabCut for still images, we developed a system to perform interactive foreground/background segmentation on a reconstructed 3D scene using an intuitive user interface [Paper] [Code]
	Face Modeling with a Commodity Depth Camera We developed a system for modeling a person's face in real-time using a low-cost depth camera, such as, the Kinect. The model is constructed by registering and integrating multiple segmented depth images of the user's head. Our method allows a user to generate a model of their face by simply moving their head in front of a fixed depth camera. [Paper] [Code]

Publications

G. P. Meyer and M. N. Do, "Real-time 3D Face Verification with a Consumer Depth Camera," in Proceedings of the 15th Conference on Computer and Robot Vision (CRV), 2018. [PDF]

G. P. Meyer, S. Alfano, and M. N. Do, "Improving Face Detection with Depth," in Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2016. [PDF]

G. P. Meyer, S. Gupta, I. Frosio, D. Reddy, and J. Kautz, "Robust Model-based 3D Head Pose Estimation," in Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2015. [PDF]

G. P. Meyer and M. N. Do, "3D GrabCut: Interactive foreground extraction for reconstructed 3D scenes," in Proceedings of the Eurographics Workshop on 3D Object Retrieval (3DOR), 2015. [PDF]

G. P. Meyer and M. N. Do, "Real-time 3D face modeling with a commodity depth camera," in IEEE International Conference on Multimedia and Expo (ICME), 2013. [PDF]

Patents

G. P. Meyer, S. Gupta, I. Frosio, D. Reddy, and J. Kautz "Model-based three-dimensional head pose estimation," US Patent 9 830 703, Nov. 28, 2017.

Q. H. Nguyen, G. P. Meyer, M. N. Do, D. Lin, and S. J. Patel, "Systems and methods for accurate user foreground video extraction," US Patent 8 818 028, Aug. 26, 2014.

Industrial Experience

Senior Software Engineer, Jan. 2017 - Present, Uber Advanced Technologies Group: Working on the perception team to develop methods for detecting objects around a self-driving vehicle by leveraging multiple sensing modalities and machine learning.

Software Engineering Intern, May 2015 - Aug. 2015, Google: Worked with the Jump team to develop a technique for capturing a room-size environment with the Jump camera.

Research Intern, Aug. 2014 - Jan. 2015, NVIDIA: Worked with the Mobile Visual Computing group at NVIDIA Research to develop a method for estimating the pose of a person's head using a low-cost depth camera.

Software Engineering Intern, May 2014 - Aug. 2014, Google: Worked with the Photo Editing team to develop techniques for editing photos using depth information.

Software Engineering Intern, May 2013 - Aug. 2013, Google: Worked with the Google Objects team at Google Research to construct a system for capturing a coarse 3D model of an object using a depth camera.

Member of Technical Staff, Jan. 2010 - May 2013, Personify Inc.: Designed and implemented real-time algorithms to segment video frames into foreground and background regions using depth and color information.