Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/308351
Title: Development of video stitching algorithms
Researcher: Krishnakumar K
Guide(s): Indira gandhi S
Keywords: Engineering and Technology
Computer Science
Computer Science Information Systems
video stitching
algorithms
University: Anna University
Completed Date: 2019
Abstract: Video stitching is a process to generate a wide field of view video from multiple over-lapping videos captured from ordinary cameras. Currently, video stitching has become a popular research topic in computer vision. Video stitching is being used in surveillance, virtual reality, augmented reality, virtual tour, games, and movies. Many algorithms for video stitching rely on locating feature points in each image, calculating feature descriptions from a region of pixels surrounding each feature point. This research work is also based on this approach for providing effective presentation of videos. In this thesis, the challenges faced while stitching the videos from multiple cameras are addressed. While stitching the videos from multiple cameras misalignment, artifacts and ghosting effects occurring due to parallax error need to be addressed. Apart from this challenge, Spatial-temporal consistency also needs to be tackled. Spatial-temporal smoothness is affected by the motion of camera as well as by the motion of objects between the frames. Videos, in general, captured by moving cameras appear shaky, which will affect temporal smoothness. The proposed work focuses on developing video stitching algorithms with Interacting Multiple Model feature tracking technique, spatial-temporal region-based method, Stationary wavelet transform based feature detection method and Multi view spatial-temporal feature method. In the Interacting Multiple Model Feature Tracking technique, the feature-based video stitching algorithms mostly rely on tracking of feature points. In this work, the spatial-temporal consistency is maintained by designing a model that accurately track the feature points and correctly newline
Pagination: xvi, 134p.
URI: http://hdl.handle.net/10603/308351
Appears in Departments:Faculty of Information and Communication Engineering

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File58.13 kBAdobe PDFView/Open
02_certificates.pdf970.66 kBAdobe PDFView/Open
03_abstracts.pdf69.38 kBAdobe PDFView/Open
04_acknowledgements.pdf66.46 kBAdobe PDFView/Open
05_contents.pdf113.06 kBAdobe PDFView/Open
06_listofabbreviations.pdf66.13 kBAdobe PDFView/Open
07_chapter1.pdf1.83 MBAdobe PDFView/Open
08_chapter2.pdf328.36 kBAdobe PDFView/Open
09_chapter3.pdf21.84 MBAdobe PDFView/Open
10_chapter4.pdf1.94 MBAdobe PDFView/Open
11_chapter5.pdf5.51 MBAdobe PDFView/Open
12_chapter6.pdf14.54 MBAdobe PDFView/Open
13_conclusion.pdf200.94 kBAdobe PDFView/Open
14_references.pdf271.13 kBAdobe PDFView/Open
15_listofpublications.pdf76.41 kBAdobe PDFView/Open
80_recommendation.pdf102.04 kBAdobe PDFView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.

Altmetric Badge: