Artistic rendering of image
Date of Issue2011
School of Electrical and Electronic Engineering
This is an experimental-based project aiming at using current computer vision technologies to extract the human out from the background automatically from portrait oil paintings and amateurish photographs. Human detection using Histogram of Oriented Gradients followed with Active Contour Model is experimented at the initial phase of this project. However, the results are not as good as expected thus an alternative is explored with the aid of Kinect depth sensor, which captures the depth image and turns out to be a feasible way to get the human contour after some processing. Also, two basic rendering tools, histogram equalization and contrast stretching are experimented on the extracted human object layer to achieve aesthetic image enhancement.
DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Final Year Project (FYP)
Nanyang Technological University