Lab Course: Building 3D Models from image sequences
Course
 Lecturer(s):
 Start: 5. März 2003
 Dates:
 Curriculum: Diploma (Graduate)
 Diploma Area: B
Description
Detailed and realistic 3D models are keyfeatures in producing highquality computer generated images. However, creating such models from scratch still remains a nontrivial, timeconsuming task and even an artistic challenge. Therefore, there is a constantly increasing need for 3D laser range scanning devices and methods, however, the accessibility to this technology is limited by ist extreme expensiveness. Thus, it would be desirable to have a method capable of generating visually satisfying reconstructions of a scene's geometry from an input of a sequence of images taken for example with a common digital camera.
To achieve this, we theoretically only need the projective camera mappings. Having these, by utilizing the inverse camera mapping, we project the two corresponding imagepoints back into the 3D space and thus obtain two rays intersecting at the location of the genuine 3D point (the socalled triangulation algorithm). By conducting this procedure for all located pairs of corresponding image points we generate a more or less dense pointcloud in the 3D space. In order to obtain a renderable model, this pointcloud has to be further processed.
In order to design and implement such a system, the following general problems have to be addressed:
 Determination of corresponding points in an image sequence
 Finding the camera matrix for each image
 Converting the retrieved data into a renderable model
After a discussion of existing and novel solutions, the goal of this practical course will be the implementation and integration of these methods into a modular system for "3Dreconstruction from images".
Slides
 Introduction to MultiViewGeometry (PDF document, 1.8 MB)
 Introduction to MultiViewGeometry (bw) (PDF document, 2.0 MB)
 Metric Reconstruction and Selfcalibration (PDF document, 598 KB)
 Metric Reconstruction and Selfcalibration (bw) (PDF document, 581 KB)
 Dense Depthmap Estimation (bw) (PDF document, 335 KB)
 Dense Depthmap Estimation (PDF document, 275 KB)
 Structure and Motion (bw) (PDF document, 141 KB)
 Structure and Motion (PDF document, 142 KB)
Assignment Sheets
Exercise 1: DenseDepthEstimation  Task 1 Assignment sheet (PDF document, 53 KB)


Exercise 2: ProjectiveReconstruction  Task 1 Assignment sheet (PDF document, 64 KB)


Exercise 3: DenseDepthEstimation  Task 2 Assignment sheet (PDF document, 62 KB) 

Exercise 4: ProjectiveReconstruction  Task 2 Assignment sheet (PDF document, 54 KB) 
Additional Documents
 Dense Disparity Estimation (PDF document, 1.3 MB)
 Determining the Epipolar Geometry (PDF document, 0.9 MB)
 Newmat  Lib Files (ZIP archive, 667 KB)
 Programming Guidelines (PDF document, 804 KB)
 Sequential Updating of Projective Structure (PDF document, 877 KB)
 Triangulation (PDF document, 179 KB)