Incremental Visual Hull Reconstruction

We propose a novel algorithm for incrementally reconstructing the visual hull in dynamic scenes by exploiting temporal consistency. Using the difference in the silhouette images between two frames we can efficiently locate the scene parts that have to be updated using ray casting. By only concentrating on the parts of the scene that have to be checked for changes, we achieve a significant speed up over traditional methods, which reconstruct each frame independently. Our method does not use any kind of scene model and works regardless of the number and shape of the objects in the scene. We test our algorithm on sequences taken with a multi-camera system and perform a detailed performance analysis showing that our method outperforms other existing methods.

[1]  Takeo Kanade,et al.  Shape-From-Silhouette Across Time Part II: Applications to Human Modeling and Markerless Motion Tracking , 2005, International Journal of Computer Vision.

[2]  Jean Ponce,et al.  Projective Visual Hulls , 2007, International Journal of Computer Vision.

[3]  Takeo Kanade,et al.  Shape-From-Silhouette Across Time Part I: Theory and Algorithms , 2005, International Journal of Computer Vision.

[4]  Adrian Hilton,et al.  Correspondence labelling for wide-timeframe free-form surface matching , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[5]  R. Zabih,et al.  Exact voxel occupancy with graph cuts , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[6]  Michael Potmesil Generating octree models of 3D objects from their silhouettes in a sequence of images , 1987, Comput. Vis. Graph. Image Process..

[7]  Richard Szeliski,et al.  Rapid octree construction from image sequences , 1993 .

[8]  Edmond Boyer,et al.  Exact polyhedral visual hulls , 2003, BMVC.

[9]  Ramakrishnan Mukundan,et al.  Ray casting for incremental voxel colouring , 2005 .

[10]  Jean-Philippe Pons,et al.  Spatio-Temporal Shape from Silhouette using Four-Dimensional Delaunay Meshing , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[11]  Nassir Navab,et al.  Efficient visual hull computation for real-time 3D reconstruction using CUDA , 2008, 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[12]  Edmond Boyer,et al.  A hybrid approach for computing visual hulls of complex objects , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[13]  BakerSimon,et al.  Shape-From-Silhouette Across Time Part II , 2005 .

[14]  Bruno Raffin,et al.  Parallel Adaptive Octree Carving for Real-time 3D Modeling , 2007, 2007 IEEE Virtual Reality Conference.

[15]  Wojciech Matusik,et al.  Articulated mesh animation from multi-view silhouettes , 2008, ACM Trans. Graph..

[16]  Oliver William Batchelor Ray Traversal for Incremental Voxel Colouring , 2006 .

[17]  John Amanatides,et al.  A Fast Voxel Traversal Algorithm for Ray Tracing , 1987, Eurographics.

[18]  Richard Szeliski,et al.  A Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[19]  Radu Horaud,et al.  Temporal Surface Tracking Using Mesh Evolution , 2008, ECCV.

[20]  Thomas Ertl,et al.  for Graphics-Hardware-based Cluster Systems , 2006 .

[21]  A. Laurentini,et al.  The Visual Hull Concept for Silhouette-Based Image Understanding , 1994, IEEE Trans. Pattern Anal. Mach. Intell..