II. 3D Computer Vision: Scene Reconstruction

     The objective of the work in this area is three-dimensional interpretation of images. A central theme is integrating the extraction and analysis of two-dimensional image structure with the enforcement of a three-dimensional model of the environment on the estimated scene structure.

II.1. Surfaces from Spatial Stereo, based on the disparity between multiple images taken from multiple cameras but at the same time.

Projects:

1. Surfaces from Binocular Spatial Stereo 2. Dense Stereo Maping Using Kernel Maximum Likelihood Estimation

II.2. 3D object modeling from multiple views and multiple depth cue.

Projects:

1. 3D surface from multiple views

II.3. 3D Surfaces, Reflectance and Illumination from Stereo, Appearence and Shading in multiple views.

Projects:

1. 3D Surfaces and Illumination from Stereo and Shading 2. Non-lambertian surface reconstruction and reflectance modeling

II.4. 3D Surface Orientation from Texture Gradient computed in a single image of a homogeneously textured surface.