Light weight Head Pose Invariant Gaze Tracking

Unconstrained remote gaze tracking using off-the-shelf cameras is a challenging problem . We propose a novel branched CNN architecture that improves the robustness of gaze classifiers to variable head pose, without increasing computational cost . We also present various procedures to effectively train our gaze network including transfer learning from the more closely related task of object viewpoint estimation and from a large high-fidelity synthetic gaze dataset, which enable our ten times faster gaze network to achieve competitive accuracy to its current state of theart direct competitor . The authors also present a number of procedures to train the gaze network using a large, high-reliant synthetic gaze datasets to achieve its competitive accuracy .

Links: PDF - Abstract

Code :

None

Keywords : gaze - network - large - accuracy - high -

Leave a Reply

Your email address will not be published. Required fields are marked *