Gaze Data for the Hollywood2 Action Recognition Data Set

2012-10-22T19:23:00Z (GMT) by David Cox Eleonora Vig Michael Dorr

Gaze data from 5 subjects watching the Hollywood2 Action Recognition Dataset while performing the action recognition task set out for the challenge.


Eleonora Vig, Michael Dorr, David Cox (2012) Space-Variant Descriptor Sampling for Action Recognition Based On Saliency and Eye Movements. European Conference on Computer Vision