In computer vision, the process of 3D reconstruction consists of rebuilding a 3D scene from a set of images. The frames (images) can be extracted from a video sequence to reconstruct the captured scene. This research study concentrates on finding the suitable frames for reconstruction and discarding the rest of the frames through the process of key frame extraction. The selection process based on the properties of the frames avoids degenerate cases, helps to improve performance, and enables the 3D reconstruction process to minimize compute time. Besides key frame extraction, I developed a browser-based 3D video player to allow users to view the captured video in three dimensions.
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder. If you wish to make any use of the work not provided for in the law, please contact: firstname.lastname@example.org