Object recognition by computer: the role of geometric constraints
Object recognition by computer: the role of geometric constraints
Three-dimensional computer vision: a geometric viewpoint
Three-dimensional computer vision: a geometric viewpoint
Machine vision
Matching feature points in image sequences through a region-based method
Computer Vision and Image Understanding
Digital Image Processing
Image-based Modeling using a Two-Step Camera Calibration Method
SIBGRAPHI '98 Proceedings of the International Symposium on Computer Graphics, Image Processing, and Vision
Object Recognition by Alignment Using Invariant Projections of Planar Surfaces
Object Recognition by Alignment Using Invariant Projections of Planar Surfaces
Non-goal scene analysis for soccer video
Neurocomputing
Camera calibration in sport event scenarios
Pattern Recognition
| Hi-index | 0.00 |
In the broadcast of sports events one can commonly see adds or logos that are not actually there - instead, they are inserted into the image, with the appropriate perspective representation, by means of specialized computer graphics hardware. Such techniques involve camera calibration and the tracking of objects in the scene. This article introduces an automatic camera calibration algorithm for a smooth sequence of images of a football (soccer) match taken in the penalty area near one of the goals. The algorithm takes special steps for the first scene in the sequence and then uses coherence to efficiently update camera parameters for the remaining images. The algorithm is capable of treating in real-time a sequence of images obtained from a TV broadcast, without requiring any specialized hardware.