Active Basis

Note: Below were experiments done at the early stage of this project. See THIS PAGE for more recent experiments where allowing rotation and flipping leads to cleaner templates.

Learning Active Basis Models from Non-aligned Images

Code for bike example

The first plot is the learned template. For each of the following pairs of images, the left is the observed image (displayed in color for visual clarity, although the algorithm works on grey level images), the right is the image superposed with the translated and deformed template.

Code for egret example

Code for cat example

Code for lion example (memory friendly version)

Code for bear example (memory friendly version)

Code for face example (with local shifting)

Code for cat example (with local shifting)

Code for bear example (with local shifting)

Old code for reproducing results in paper: bike horse crane camel cat

Back to active basis homepage