Segmentation of spatiotemporal data based on user gaze data

摘要

A segmentation task is specified to a user, and gaze data generated by monitoring eye movements of the user viewing spatiotemporal data as a plurality of frames is received. The gaze data includes fixation locations based on the user's gaze throughout the frames. A first frame and a second frame of the frames are selected based on the fixation locations. Segmentation is performed on the first and second frames to segment first and second objects, respectively, from the first and second frames based on a region of interest associated with the first and second frames, the region of interest corresponding to a location of one of the fixation locations. A determination is made as to whether the first and second objects are relevant to the segmentation task, and if so, association data to associate the first object with the second object when the first and second objects is generated.