Determining a depth map using a dual photometric stereo
International Journal of Robotics Research
Simulated annealing: theory and applications
Simulated annealing: theory and applications
Integration of visual modules: an extension of the Marr paradigm
Integration of visual modules: an extension of the Marr paradigm
Surface descriptions from stereo and shading
Shape from shading
Reconstructing a 3-D depth map from one or more images
CVGIP: Image Understanding
IEEE Transactions on Computers - Special issue on artificial neural networks
Cooperative Vision Integration Through Data-Parallel Neural Computations
IEEE Transactions on Computers - Special issue on artificial neural networks
Surface reconstruction from stereoscopy and “shape from shading” in SEM images
Machine Vision and Applications - Special issue: Three-dimensional microscopy
Numerical recipes in C (2nd ed.): the art of scientific computing
Numerical recipes in C (2nd ed.): the art of scientific computing
Using 3-dimensional meshes to combine image-based and geometry-based constraints
ECCV '94 Proceedings of the third European conference on Computer Vision (Vol. II)
From Images to Surfaces: A Computational Study of the Human Early Visual System
From Images to Surfaces: A Computational Study of the Human Early Visual System
Efficiently combining positions and normals for precise 3D geometry
ACM SIGGRAPH 2005 Papers
3D shape recovery by the use of single image plus simple pattern illumination
ISVC'07 Proceedings of the 3rd international conference on Advances in visual computing - Volume Part I
A 3D Imaging Framework Based on High-Resolution Photometric-Stereo and Low-Resolution Depth
International Journal of Computer Vision
Hi-index | 0.00 |
In the domain of 3-D scene reconstruction this work presents the cooperation of shape from shading and stereo vision and demonstrates how to overcome a certain number of previously encountered problems. The problems of application assumptions, regularization terms and simplifications of physical models, used to overcome the problem of the modules of being ill-posed, are solved by the concept of integrating complementary knowledge of the physical world into one system. The problems due to the use of non-optimal resolution methods and too long parameter lists when the modules are integrated in a homogeneous system, are solved by the introduction of a cooperation concept for heterogeneous systems. The problem of error propagation from stereo vision to shape from shading. when only the initial and border conditions are used for the cooperation, is solved by the introduction of simultaneous constraints from both modules on all image points. The shape from shading problems of using too simple physical models for real scenes and inconsistent physical models with stereo vision are overcome by the introduction of more complex physical models: perspective projection, point light sources and Phong's reflection model. The stereo vision problem caused by the lack of a global quality constraint when correlation is used as resolution method, is solved by using simulated annealing. The stereo vision problem arising from the use of the gray-levels for the resemblance constraint and so assuming lambertian surfaces, is solved by using the photometric characteristics from shape from shading instead.