Reconstruction and prediction from three images of uncalibrated cameras
Selected papers from the 9th Scandinavian conference on Image analysis : theory and applications of image analysis II: theory and applications of image analysis II
Theory of Reconstruction from Image Motion
Theory of Reconstruction from Image Motion
A Common Framework for Kinetic Depth, Reconstruction and Motion for Deformable Objects
ECCV '94 Proceedings of the Third European Conference-Volume II on Computer Vision - Volume II
Reconstruction from image sequences by means of relative depths
ICCV '95 Proceedings of the Fifth International Conference on Computer Vision
Critical Motions for Auto-Calibration When Some Intrinsic Parameters Can Vary
Journal of Mathematical Imaging and Vision
An Efficient Solution to the Five-Point Relative Pose Problem
IEEE Transactions on Pattern Analysis and Machine Intelligence
Some Results on Minimal Euclidean Reconstruction from Four Points
Journal of Mathematical Imaging and Vision
The Five Points Pose Problem: A New and Accurate Solution Adapted to Any Geometric Configuration
PSIVT '09 Proceedings of the 3rd Pacific Rim Symposium on Advances in Image and Video Technology
ACM Transactions on Graphics (TOG)
A simple solution to the six-point two-view focal-length problem
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part IV
| Hi-index | 0.00 |
This paper deals with the problem of reconstructing the locations offive points in space from two different images taken by calibratedcameras. Equivalently, the problem can be formulated as finding thepossible relative locations and orientations, in three-dimensionalEuclidean space, of two labeled stars, offive lines each, such that corresponding lines intersect.The problem was first treated by Kruppa more than 50 years ago. He found that there were at most eleven solutions. Later Demazure and alsoMaybank showed that there were actually ten solutions. In this articlewill be given another proof of this theorem based on a different parameterisation of the problem neither using the epipoles nor theessential matrix. This is within the same point of view as directstructure recovery in the uncalibrated case. Instead of the essentialmatrix we use the kinetic depth vectors, which hasshown to be were useful in the uncalibrated case. We will alsopresent an algorithm that in most cases calculates the ten differentsolutions, although some may be complex and some may not be physicallyrealisable. The algorithm is based on a homotopy method and trackssolutions on the so called Chasles‘ manifold. One of the majorcontributions of this paper is to bridge the gap between reconstructionmethods for calibrated and uncalibrated cameras. Furthermore, we show thatthe twisted pair solutions are natural in this context because the kineticdepths are the same for both components.