Abstract: The present disclosure provides a system and method for predicting drowsiness of a subject. During facial landmark tracking of a received stream of images, feature extraction is split into two components, viz., landmark independent and landmark dependant features. Precomputing the landmark independent features during a first stage of a first cascade reduces the computational requirements of the approach without affecting accuracy of detection of landmark features. A second multi-stage cascade is used to distinguish and extract eye images from the facial landmarks. At the end of the second multi-stage cascade, the eye landmarks are again distinguished between open eyes and closed eyes using a third multi-stage cascade. Progressively, the subsequent stages of the third multi-stage cascade discards open eyes, and eventually detects closed eyes accurately. The occurrence of closed eye images and eye height variations in the received stream of images can enable determination of drowsiness of the subject.

Abstract: A computer-implemented method for tracking an object in a video including, in a first video frame: filtering the first video frame to delineate foreground from background and to select from the foreground an object to be tracked, said object identified at least by contour, center and color; constructing a first virtual bounding box minimally enclosing said object; using a tracking algorithm to establish a target location for the object in the second video frame based on similarity to the object identified in the first video frame; constructing a second virtual bounding box minimally enclosing the target shape at the target location; extending the second virtual bounding box; filtering the second video frame to delineate foreground from background using weighted values propagated from at least the first frame; scrutinizing the output of the second video frame by matching at least shape and color of the target shape at the target location.