Real-time stereo-based view synthesis algorithms: A unified framework and evaluation on commodity GPUs

Rogmans, Sammy; Lu, Jiangbo; Bekaert, Philippe; Lafruit, Gauthier

doi:10.1016/j.image.2008.10.005

Cited by 30 publications

(25 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Those techniques proceed in general in 3 stages [1,2]: the estimation of correspondences between the novel and the reference views, the estimation of the contribution (or weight) of each view in the final view, and the rendering method. The stereo correspondence estimation problem has been largely explored [3], but few works have formally studied which is the correct leverage between the contributions of each view.…”

Section: Introductionmentioning

confidence: 99%

Viewpoint interpolation: Direct and variational methods

Pujades

Devernay

2014

2014 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

We address the topic of novel view synthesis from a stereoscopic pair of images. Most current techniques are composed of 3 stages: reconstruction of correspondences between the views, estimation of the per-pixel or per-image weight of each view in the final view, and rendering. Rendering methods can be classified into direct methods, which define the final image as a function of the original images, and variational methods, where the synthesized image is expressed as the solution minimising an energy. The existing literature is mainly focused on the correspondence issue, but few publications address the questions of which weights or blending factors are better, or which rendering method should be used. In this paper, we experiment different combinations of blending factors and rendering methods, in order to evaluate the influence of these two factors on the final image quality.Index Terms-Viewpoint interpolation, image-based rendering, blending factors, variational method.

show abstract

Section: Introductionmentioning

confidence: 99%

Viewpoint interpolation: Direct and variational methods

Pujades

Devernay

2014

2014 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

show abstract

“…Methods to synthesize new views often consist of two basic steps [19]. First a disparity estimation and then the synthesis of the new view.…”

Section: Related Workmentioning

confidence: 99%

Making of Who Cares? HD Stereoscopic Free Viewpoint Video

Lipski

Klose

Ruhl

et al. 2011

2011 Conference for Visual Media Production

View full text Add to dashboard Cite

We present a detailed blueprint of our stereoscopic freeviewpoint video system.Using unsynchronized footage as input, we can render virtual camera paths in the post-production stage.The movement of the virtual camera also extends to the temporal domain, so that slow-motion and freeze-and-rotate shots are possible. As a proof-of-concept, a full length stereoscopic HD music video has been produced using our approach.

show abstract

“…When the shooting geometry is constrained, or when the same stereoscopic movie must be displayed with different viewing geometries, the shape distortions may be reduced by new view synthesis techniques [4,5,6], which usually involve three steps: computing the stereo disparity, computing a disparity-dependent 2-D mapping from the original stereo pair to the synthesized views, and finally composing the synthesized views from the original images, the disparity, and the disparity-dependent mapping.…”

Section: New View Synthesismentioning

confidence: 99%

New view synthesis for stereo cinema by hybrid disparity remapping

Devernay

Duchêne

2010

2010 IEEE International Conference on Image Processing

View full text Add to dashboard Cite

The 3-D shape perceived from viewing a stereoscopic movie depends on the viewing conditions, most notably on the screen size and distance, and depth and size distortions appear because of the differences between the shooting and viewing geometries. When the shooting geometry is constrained, or when the same stereoscopic movie must be displayed with different viewing geometries (e.g. in a movie theater and on a 3DTV), these depth distortions may be reduced by new view synthesis techniques. They usually involve three steps: computing the stereo disparity, computing a disparity-dependent 2-D mapping from the original stereo pair to the synthesized views, and finally composing the synthesized views. In this paper, we compare different disparity-dependent mappings in terms of perceived shape distortion and alteration of the images, and we propose a hybrid mapping which does not distort depth and minimizes modifications of the image content.

show abstract

Real-time stereo-based view synthesis algorithms: A unified framework and evaluation on commodity GPUs

Cited by 30 publications

References 30 publications

Viewpoint interpolation: Direct and variational methods

Viewpoint interpolation: Direct and variational methods

Making of Who Cares? HD Stereoscopic Free Viewpoint Video

New view synthesis for stereo cinema by hybrid disparity remapping

Contact Info

Product

Resources

About