Reduced Depth and Visual Hulls of Complex 3D Scenes

Bogomjakov, Alexander; Gotsman, Craig

doi:10.1111/j.1467-8659.2008.01114.x

Cited by 8 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…11). Compared with implicit assumption used in previous geometrical approaches like "Real Shape Hulls" (RSH) [4], "Safe Hulls" (SH) [3] or "Reduced Visual Hull" (RV H) [9], our assumption is less constraining.…”

Section: Assumption Discussionmentioning

confidence: 98%

“…This approach can be unreliable with objects of heterogeneous sizes and motions. In [9], Bogomjakov and Gotsman present an algorithm that removes ghost-components from V H, using correspondences knowledge between silhouettes associated to each object. Out of color matching instabilities with unknown illumination, silhouette correspondences impose that each object is not, even partially, occluded from at least two views.…”

Section: Previous Workmentioning

confidence: 99%

See 1 more Smart Citation

Towards Removing Ghost-Components from Visual-Hull Estimations

Michoud

Guillou

Bouakaz

et al. 2009

2009 Fifth International Conference on Image and Graphics

View full text Add to dashboard Cite

Shape from Silhouette (SFS) are methods used to estimate the 3D shape of objects from their silhouettes. The reconstructed shape, also called Visual Hull (VH), is the maximum volume which yields the silhouettes. As this volume is an upper bound, it contains objects and artifacts. Ghost connected components coming from artifacts are parts empty of real object. They provide non-pertinent data that interfere with many applications of SFS: motion capture, free-viewpoint rendering, etc. The more the object's number increases, the more ghost-components appears. The problem is even worst with a small number of cameras.In this paper, we address the issue of detecting and removing ghost-components from any Visual Hull. Our method only needs cameras' calibration and silhouette data. A formal description and several practical experimentation, show the efficiency of our approach. A comparison with previous methods underlines more accurate results, even using few cameras.

show abstract

Section: Assumption Discussionmentioning

confidence: 98%

Section: Previous Workmentioning

confidence: 99%