Abstract
Given a video signal, we generate an alpha matte based on the chromakey information. The computation is done in interactive-time using pixel shaders. To accomplish this, we use Principle Components Analysis to generate a linear transformation matrix where the resulting color triplets Euclidean distance is directly related to the probability that the color exists in the chromakey spectrum. The result of this process is a trimap of the video signals opacity. To solve the alpha matte from the trimap, we minimize an energy function constrained by the trimap with gradient descent. This energy function is based on the least-squared error of overlapping neighborhoods around each pixel and is independent of the background or foreground color.
Document Type
Patent
Patent Number
US 8,477,149
Application Serial Number
12/752,238
Issue Date
7-2-2013
Current Assignee
UCFRF
Assignee at Issuance
UCFRF
College
College of Engineering and Computer Science (CECS)
Department
Electrical Engineering & Computer Science - CS Division
Allowance Date
3-5-2013
Filing Date
4-1-2010
Assignee at Filing
UCFRF
Filing Type
Nonprovisional Application Record
Donated
no
Recommended Citation
Hughes, Charles; Beato, Nicholas; Colbert, Mark; Yamazawa, Kazumasa; and Zhang, Yunjun, "Real-Time Chromakey Matting Using Image Statistics" (2013). UCF Patents. 490.
https://stars.library.ucf.edu/patents/490