Flexible Bump Map Capture From Video James A
Flexible Bump Map Capture From Video James A. Paterson and Andrew W. Fitzgibbon University of Oxford {jamie, awf}@robots. ox. ac. uk Bump Mapping The Algorithm Calibration • Requirement: Avoid physical measurement …is a generic term grouping techniques which use 2 D texture to apply 3 D detail to a surface. This includes Displacement mapping and our primary interest Normal mapping. We present an improved method for normal map acquisition from real world objects. • Use set of calibration images of light reflected in mirror, intersect ray on which light must lie in at least two images to find relative position Input Images – A video sequence of the object being moved in front of a calibrated camera-light rig, or equivalently a portable camera-light rig moved around whilst viewing a static object Calibration geometry Standard method Equivalent new method(s) Images used in calibration Fronto-parallel Images – By tracking markers on the object we can compute relative camera-light position, and re-render the object • Measurement error makes closed from solution inaccurate. Use a using an inverse perspective projection. This is done via 2 D non-linear optimisation scheme to optimise camera positions, focal length and light position, minimizing reprojection error: homography for planar markers. • Standard normal map capture rig is accurate, but bulky. • Requires static camera and multiple static lights. • Proposed new version is portable, only requires one light • Relative camera-light position is calculated, and must be constant Calculating a surface normal map To find the normal at a particular point on the surface, we observe the intensity of that point under different lighting conditions. We model lighting effects via the standard Lambertian equations. Normal map capture requires at least 3 different intensity values for a point on the surface, captured with different known light directions. Then we can find the normal at that point as follows: Sparse Normal Map Images – We can now compute surface normals as outlined previously. Here we see a sparse set of normals superimposed on input video frames Current issues and possible extensions • The Lambertian lighting approximation is not perfect, but by using more than 3 images we can avoid irregular intensity values caused by specular highlights and self shadowing. Determining exactly which pixels from which images to include in the computation is non-trivial, but we would hope to maximise on the range of light directions used in calculating each normal. • Currently only planar surfaces can be captured, however by Dense Normal Map Images – A higher density grid superimposed improving the geometrical model of the subject we hope to extend on a longer sequence of input images as a bump mapped the process to highly curved surfaces. quadrilateral. Non-Lambertian lighting conditions (specular • The positions of the markers on the subject must be accurately highlights, self shadowing) lead to occasional spurious results. known, however in e. g. archive footage this information will not be available. Given sufficient images we can include marker positions Limitation: Planar surfaces only at the moment! in the non-linear optimisation, removing this restriction
- Slides: 1