Vittorio Ferrari, Frederic Jurie, Cordelia Schmid

Overview

Matlab source code for learning shape models of object categories from unsegmented training images, and for localizing outlines of new category members in cluttered test images. Although the training stage requires the objects to be annotated only by bounding-boxes, at test time the method can localize objects up to their boundaries.

This release covers the CVPR 2007 [1] and IJCV 2010 [2] papers.

Downloads

History

New in V 1.3

• now exactly identical to what described in our papers and can reproduce exactly the same performance plots
• includes DR/FPPI performance curves from our papers in matlab .fig format
• won’t crash on images with zero detections
• fully compatible with output of kAS detector package

References

1. Accurate Object Detection with Deformable Shape Models Learnt from Images
   V. Ferrari, F. Jurie, and C. Schmid,
   In Computer Vision and Pattern Recognition (CVPR), 2007.
2. From Images to Shape Models for Object Detection
   V. Ferrari, F. Jurie, and C. Schmid,
   In International Journal of Computer Vision (IJCV), 2010.