PET
PASCAL Animal Classes Eye Tracking Database
We present the PASCAL animal classes Eye Tracking database (PET). PET comprises eye movement recordings compiled for the bird, cat, cow, dog, horse, and sheep training+validation (or trainval) sets from the PASCAL VOC 2012 image database. These six animal-centric classes were chosen for the following reasons:
- Animal classes such as cats, dogs and birds are particularly difficult to detect using traditional supervised learning methods (e.g. deformable parts model) owing to large intrinsic shape and textural variations.
- It would be highly beneficial to incorporate human knowledge to train object detectors for these classes, as many psychophysical studies have noted our tendency to instantaneously detect animals (which are both predators and prey).
A salient aspect of the PET dataset is that it contains eye movements recorded under both the free-viewing (no task) and visual search (task-based) paradigms. In all, eye movements were recorded from a total of 40 users, who viewed each image for 2 seconds, so that four gaze patterns are available per image and condition.
The dataset is released under a creative commons license and can be downloaded here: pet.zip (343MB). The zip-file is encrypted; please email us for the password.
More details about the database can be found in the following paper:
PET: An eye-tracking dataset for animal-centric PASCAL object classes. (oral presentation)
IEEE International Conference on Multimedia and Expo (ICME), Torino, Italy, June 29-July 3, 2015.
Please cite the above paper if you use the PET dataset.