Abstract
This paper introduces a representation scheme for images and video sequences using nonuniform samples embedded in a mesh structure. It describes a video sequence by the nodal positions and colors in a starting frame, followed by the nodal displacements in the following frames. The nodal points are more densely distributed in regions coniaining inieresting features such as edges and corners; and are dynamically updated to follow the same fealures in successive frames. They are determined automatically by maximizing feature (e.g, gradient) magnitudes at nodal poinis, while minimizing interpolation errors within individual elements, and matching errors between corresponding elements. In order to avoid the mesh elements to become overly deformed, a penalty term is also incorporated, which measures the irregularity of the mesh structure. The noiions of shape functions and master elements commonly used in the finite element method have been employed lo simplify the numerical calculation of the energy functions and their gradients. The proposed representation is motivated by the active contour or snake model proposed by Kass, Within and Terzopoulos. The current representation retains the salient merit of the original model as a feature tracker brsed on local and collective information, while facilitating more accurate image interpolation and prediction.
Original language | English (US) |
---|---|
Pages (from-to) | 1558-1569 |
Number of pages | 12 |
Journal | Proceedings of SPIE - The International Society for Optical Engineering |
Volume | 2094 |
DOIs | |
State | Published - 1993 |
Event | Visual Communications and Image Processing 1993 - Cambridge, MA, United States Duration: Nov 7 1993 → Nov 7 1993 |
ASJC Scopus subject areas
- Electronic, Optical and Magnetic Materials
- Condensed Matter Physics
- Computer Science Applications
- Applied Mathematics
- Electrical and Electronic Engineering