WaldBoost detector - Object Detection using Boosted Features

Class for object detection using WaldBoost from [Sochman05]. Uses a Waldboost cascade and local binary patterns computed as integral features for 2D object detection.

Example

The basic usage is the following:

% train
detector = cv.WBDetector();
detector.train('/path/to/pos/', '/path/to/neg/');
detector.write('model.xml');

% detect
detector = cv.WBDetector();
detector.read('model.xml');
img = cv.imread('image.png', 'Grayscale',true);
[bboxes,conf] = detector.detect(img);
for i=1:numel(bboxes)
    img = cv.rectangle(img, bboxes{i}, 'Color',[0 255 0]);
end
imshow(img)

References

[Sochman05]:

J. Sochman and J. Matas. "WaldBoost - Learning for Time Constrained Sequential Detection", IEEE Conference on CVPR 2005, Vol 2, p. 150-156 PDF