If I feed an imaged and its mirrored version to the matlab saliency toolbox, I would expect to get two saliency map which are identical except for the mirroring. However, the saliency toolbox returns me two different saliency maps. Does this reveal a bug in the algorithm, or is it correct that it results in two different maps? I implemented the Itti et al's 1998 description of the algorithm. This implementation does give me two identical salieny maps for the two mirrored versions.
Can anyone help me on this?
Cheers,
Gert Kootstra
Artificial Intelligence
University of Groningen, the Netherlands
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Welcome to the SaliencyToolbox! In this forum, please post help requests for working with the SaliencyToolbox. Enjoy!
Dirk Walther
Hi,
If I feed an imaged and its mirrored version to the matlab saliency toolbox, I would expect to get two saliency map which are identical except for the mirroring. However, the saliency toolbox returns me two different saliency maps. Does this reveal a bug in the algorithm, or is it correct that it results in two different maps? I implemented the Itti et al's 1998 description of the algorithm. This implementation does give me two identical salieny maps for the two mirrored versions.
Can anyone help me on this?
Cheers,
Gert Kootstra
Artificial Intelligence
University of Groningen, the Netherlands