Image/video analysis in computer vision

Show simple item record Sonane, Bhoomika 2017-03-23T05:39:17Z 2017-03-23T05:39:17Z 2016
dc.identifier.citation Sonane, B. (2016). Image/video analysis in computer vision. Indian Institute of Technology Gandhinagar, pp. 52. (Acc No: T00133) en_US
dc.description.abstract Image matting is the process to extract an accurate foreground from a given image. The estimated foreground is used to be composited with other arbitrary background. When the matte is taken from a natural scene, rather than from an image where the background is already known, then the matting is called natural image matting. Techniques where scene is shot in front of a blue/green background is known as blue screen or chroma key techniques. The thesis deals with the applications of natural image matting in de-blurring the partially defocused images and in video matting. Image de-blurring is a technique to deconvolve a given blurred image with a digital filter to recover a good approximation of the original high quality image with comparatively less amount of blur present. Several models are used to estimate a good blur kernel (digital filter) corresponding to the blur amount present in the given image and the original image is reconstructed using deconvolution techniques. Video matting is an extension of image matting and is used to extract the foreground matte from an arbitrary background of every frame in a video sequence. An automatic scribbling approach based on the relative motion of the foreground object with respect to the background in a video is introduced for video matting. The proposed scribble propagation and the subsequent isolation of foreground and background is much more intuitive than the conventional trimap propagation approach used for video matting. Alpha maps are propagated according to the optical flow estimated from the consecutive frames to get a preliminary estimate of the foreground and background in the following frame. Accurate scribbles are placed near the boundary of the foreground region. This is required for refining the scribbled image with the help of morphological operations. We show that a high quality matte of foreground object can be obtained using state of the art image matting technique. We show that the results obtained using the proposed method are accurate and comparable with that of other state of the art video matting techniques. en_US
dc.description.statementofresponsibility by Bhoomika Sonane
dc.format.extent 52 p,: col., ill; 30 cm+.
dc.language.iso en_US en_US
dc.publisher Indian Institute of Technology Gandhinagar en_US
dc.title Image/video analysis in computer vision en_US
dc.type Thesis en_US
dc.contributor.department Electrical Engineering M.Tech.

Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search Digital Repository


My Account