Do weak readers in rural India automatically read same language subtitles on Bollywood films? An eye gaze analysis.

Same Language Subtitling (SLS) of audio-visual content on mainstream TV entertainment to improve mass reading literacy was first conceived and piloted in India. SLS is now being scaled up nationally to ensure that the reading skills of one billion TV viewers, including 600 million weak readers, remain on a lifelong pathway to practice, progress, and proficiency. Will weak readers ignore or try to read along with SLS? Our eye-tracking study investigates this question with 136 weak readers drawn from a remote village in Rajasthan state by showing them popular Hindi film clips of dialog and songs, with and without SLS.

I2DNet - Design and Real-Time Evaluation of Appearance-based gaze estimation system.

Gaze estimation problem can be addressed using either model-based or appearance-based approaches. Model-based approaches rely on features extracted from eye images to fit a 3D eye-ball model to obtain gaze point estimate while appearance-based methods attempt to directly map captured eye images to gaze point without any handcrafted features. Recently, availability of large datasets and novel deep learning techniques made appearance-based methods achieve superior accuracy than model-based approaches.