Abstract
With the Depth-image-based rendering (DIBR) technique, virtual views can be drawn at any position between two reference viewpoints. However, DIBR algorithm may introduce various distortions in the synthesized images, such as holes, artifact, and stretching. This paper proposes a quality evaluation method for DIBR-synthesized images by analyzing the holes and expanded regions. First, the holes regions are obtained with the watershed algorithm and used to evaluate the quality of the DIBR-synthesized images. Then, the synthesized image is compared with the reference image in the sub-channel, and the number of different pixel values is used to assess the global quality of synthesized images. Finally, the two measures are pooled to calculate the overall quality score of the DIBR synthesized image. Experimental results show that our scheme can better predict the image quality and performs better in terms of implementation time compared with other schemes.
Highlights
In recent years, stereoscopic multi-view video has developed rapidly and occupied an important position in 2Dto-3D video conversion system and in free viewpoint TV (FTV) [1], [2]
An effective evaluation method for DIBR-synthesized images is beneficial to improve the quality of the synthesized virtual image and it is important for the development of 3D technology
For evaluating the expanded regions, we measure the difference between the synthesized image and the reference image in each channel, and we calculate the area of the expanded regions
Summary
Stereoscopic multi-view video has developed rapidly and occupied an important position in 2Dto-3D video conversion system and in free viewpoint TV (FTV) [1], [2]. Stereoscopic multi-view video gives the audience an immersive experience and allows viewers to experience the beauty of the world [3]. It needs a lot of space and bandwidth to store and transfer data respectively. In order to reduce the space and bandwidth, the depth-imagebased-rendering (DIBR) technique was proposed, which used the depth map information and the original reference image to obtain the virtual viewpoint image. An effective evaluation method for DIBR-synthesized images is beneficial to improve the quality of the synthesized virtual image and it is important for the development of 3D technology
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
