MS Computer Science & Information Systems

Computer vision

Advanced Study: Computer Vision

As part of the degree requirements, students complete a research project or theses in an area of interest to them. This enhances their expertise in an area.

Computer Vision is is a field that attempts to have computers gain high-level understanding from digital images or videos. Computer Vision seeks to automate tasks that humans do easily. [7][8][9]

The research being conducted at UNCW is about matching and registering different types of 2D images and 3D data. Collecting point clouds and depth information with different types of 3D sensors is an integral part of the work[1]. Photographs, videos, and 3D information provide complementary information about the same scene. Photos capture high resolution textures, color, and lighting details. Videos recode moving objects and people in an area. 3D scans can quickly and accurately provide the real-world layout of buildings, objects, and foliage.

Fusing all of this information can be used for a variety of exciting applications. Video sets can be aligned with a 3D model and played in a single multi-dimensional environment[2][3]. Photographs taken over extended periods of time can be visualized simultaneously, unveiling changes in architecture, foliage, and human activity. Depth information for 2D photos and videos can be computed and used to transform the imagery into 3D pictures and movies[4]. Real-world measurements of buildings and natural sites can be provided to users through simple interactions with traditional photographs.

However, registering this type of multi-modality data is a very challenging task given the repetition and ambiguity that often occur in man-made scenes as well as the variety of properties different renderings of the same subject can possess[5][6]. Image sets collected over a period of time during which the lighting conditions and scene content may have changed, different artistic renderings, varying sensor types, focal lengths, and exposure values can all contribute to visual variations in data sets. So far, researchers at UNCW have addressed these obstacles by developing methods for visualizing different types of data so that the common features visible across dimensions and sensor types are emphasized. It is then much easier to describe and match features between imagery that can be used for registration.

Expert Faculty

Students who want to specialize in computer vision can study independently with faculty who actively research and publish in the field, and who also have significant hands-on knowledge of computer vision. Several research topics in this area include: 3D reconstruction from different types of images;registration of photographs, videos, medical images; image/camera localization; object recognition; scene analysis; uses of 3D sensors such as RGBD, infrared, and LIDAR; robotics; tracking of people, animals, cars, and other objects in different types of image and video sets; and augmented reality.

  • Dr. Brittany Morago, Assistant Professor of Computer Science
    1. Morago, B., Bui, G., and Duan, Y. Integrating LIDAR Range Scans and Photographs with Temporal Changes, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, June 2014, pp. 732-737.
    2. Bui, G., Morago, B., Antequera, R., Nguyen, T., Calyam, P., and Duan, Y., “LIDAR-based virtual environment study for disaster response scenarios,” in IEEE Integrated Management, May 2015, pp. 790-793.
    3. Gargees, R., Morago, B., Pelapur, R., Chemodanov, D., Calyam, P., Oraibi, Z., Duan Y., Seetharaman, G., and Palaniappan, K. Incident-Supporting Visual Cloud Computing Utilizing Software-Defined Networking”, IEEE Transactions on Circuits Systems for Video Technologies, May 2016.
    4. Bui, G., Morago, B., Le, T., Karsch, K., Zheyu, L., and Duan, Y. Integrating Videos with LIDAR Scans for Virtual Reality. In IEEE Virtual Reality, Greenville, South Carolina, March 2016.
    5. Morago, B., Bui, G., and Duan, Y. “2D Matching Using Repetitive and Salient Features in Architectural Images”, IEEE Transactions on Image Processing, August 2016.
    6. Morago, B., Bui, G., and Duan, Y. “An ensemble approach to image matching using contextual features”, IEEE Transactions on Image Processing, 24(11):pp. 4474-4487, 2015.

Other References

  1. Dana H. Ballard; Christopher M. Brown (1982). Computer Vision. Prentice Hall. ISBN 0-13-165316-4.
  2. Huang, T. (1996-11-19). Vandoni, Carlo, E, ed. Computer Vision : Evolution And Promise (PDF). 19th CERN School of Computing. Geneva: CERN. pp. 21–25. doi:10.5170/CERN-1996-008.21. ISBN 978-9290830955.
  3. Milan Sonka; Vaclav Hlavac; Roger Boyle (2008). Image Processing, Analysis, and Machine Vision. Thomson. ISBN 0-495-08252-X.