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: Disclosed herein is a method for global motion estimation for video stabilization. The method enables selecting a video frame from a video stream. The method further enables downscaling the video frames by factor close to 2 in a two dimensional range, dividing the downscaled video frame into a plurality of macroblocks, performing motion estimation for the macroblocks to generate a set of local motion vectors. Further, the method enables selecting macroblocks representing global motion vectors from the set of local motion vectors, computing a single global motion vector for the selected macroblocks and determining occurrence of at least one of: scene change, illumination change or crossing object and modifying the single global vector to compensate for errors induced due to occurrence of at least one of: scene change, illumination change or crossing object.

Abstract: Disclosed herein is a method for global motion estimation for video stabilization. The method enables selecting a video frame from a video stream. The method further enables downscaling the video frames by factor close to 2 in a two dimensional range, dividing the downscaled video frame into a plurality of macroblocks, performing motion estimation for the macroblocks to generate a set of local motion vectors. Further, the method enables selecting macroblocks representing global motion vectors from the set of local motion vectors, computing a single global motion vector for the selected macroblocks and determining occurrence of at least one of: scene change, illumination change or crossing object and modifying the single global vector to compensate for errors induced due to occurrence of at least one of: scene change, illumination change or crossing object.