Abstract / Description of output
Common 3D acquisition techniques, such as laser scanning and stereo capture, are realistically only 2.5D in nature. Here we consider the automated completion of hidden or missing portions in 3D scenes originally acquired from 2.5D (or 3D) capture. We propose an approach based on the non-parametric propagation of available scene knowledge from the known (visible) scene areas to these unknown (invisible) 3D regions in conjunction with an initial underlying geometric surface completion.
Original language | English |
---|---|
Pages (from-to) | 2249-2255 |
Number of pages | 7 |
Journal | IEEE Transactions on Pattern Analysis and Machine Intelligence |
Volume | 30 |
Issue number | 12 |
DOIs | |
Publication status | Published - Dec 2008 |
Keywords / Materials (for Non-textual outputs)
- Computational Geometry and Object Modeling
- Computer Graphics
- Computer Vision
- Image Processing and Computer Vision
- Range Data
- Shape
- Surface Fitting