Circular Rectifiction of 3D Video and Efficient Modification of 3D-HEVC

06/09/2023
by   Jarosław Samelak, et al.
0

Video acquired from multiple cameras located along a line is often rectified to video virtually obtained from cameras with ideally parallel optical axes collocated on a single plane and principal points on a line. Such an approach simplifies video processing including depth estimation and compression. Nowadays, for many application video, like virtual reality or virtual navigation, the content is often acquired by cameras located nearly on a circle or on a part of that. Therefore, we introduce new operation of circular rectification that results in multiview video virtually obtained from cameras located on an ideal arc and with optical axes that are collocated on a single plane and they intersect in a single point. For the circularly rectified video, depth estimation and compression are simplified. The standard 3DHEVC codec was designed for rectified video and its efficiency is limited for video acquired from cameras located on an arc. Therefore, we developed a 3-D HEVC codec modified in order to compress efficiently circularly rectified video. The experiments demonstrate its better performance than for the standard 3D-HEVC codec.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/07/2023

Depth Estimation Analysis of Orthogonally Divergent Fisheye Cameras with Distortion Removal

Stereo vision systems have become popular in computer vision application...
research
10/14/2020

Unsupervised Learning of Depth and Ego-Motion from Cylindrical Panoramic Video with Applications for Virtual Reality

We introduce a convolutional neural network model for unsupervised learn...
research
03/02/2022

Omnidirectional MediA Format (OMAF): Toolbox for Virtual Reality Services

This paper provides an overview of the Omnidirectional Media Format (OMA...
research
07/07/2022

Visual-Assisted Sound Source Depth Estimation in the Wild

Depth estimation enables a wide variety of 3D applications, such as robo...
research
09/29/2019

SteReFo: Efficient Image Refocusing with Stereo Vision

Whether to attract viewer attention to a particular object, give the imp...
research
01/04/2019

Unsupervised Learning of Depth and Ego-Motion from Panoramic Video

We introduce a convolutional neural network model for unsupervised learn...
research
11/29/2022

A Geometric Model for Polarization Imaging on Projective Cameras

The vast majority of Shape-from-Polarization (SfP) methods work under th...

Please sign up or login with your details

Forgot password? Click here to reset