Building detection and description from a single intensity image
Computer Vision and Image Understanding
The role of color attributes and similarity grouping in 3-D building reconstruction
Computer Vision and Image Understanding
Automatic object extraction from aerial imagery—a survey focusing on buildings
Computer Vision and Image Understanding
Detection and Modeling of Buildings from Multiple Aerial Images
IEEE Transactions on Pattern Analysis and Machine Intelligence
Shape Matching and Object Recognition Using Shape Contexts
IEEE Transactions on Pattern Analysis and Machine Intelligence
Edge Flow: A Framework of Boundary Detection and Image Segmentation
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Learning a kernel function for classification with small training samples
ICML '06 Proceedings of the 23rd international conference on Machine learning
Object detection combining recognition and segmentation
ACCV'07 Proceedings of the 8th Asian conference on Computer vision - Volume Part I
Man-made structure detection in natural images using a causal multiscale random field
CVPR'03 Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition
Delineating buildings by grouping lines with MRFs
IEEE Transactions on Image Processing
Hi-index | 0.00 |
In this paper, a new building detection scheme using semi-supervised edge learning is proposed. This scheme utilizes a feature based on edge flow to delineate the patterns of sharp contrast at the edges of building. The contrast patterns with their distribution in the features space based on similarity metric provide discriminative evidences for the building detection. By the extended kernelBoosting, the semi-supervised edge learning, a number of Gaussian Mixture Models (GMMs) are computed and optimized to model the local distribution of contrast patterns according to their similarity. The `weak kernel' hypotheses are then generated from these optimized Gaussian Mixture Models. The final kernel is defined by accumulating a weighted linear combination of such "weak kernel". The kernel function can then be used for classification with kernel SVM. Experiments show that this scheme is capable of achieving both low false positive rate and low false negative rate with only a few training examples and that this method can be generalized to many object classes.