Optimized volumetric visual hull reconstruction method based on CUDA

Volumetric visual hull reconstruction is easy to be implemented without complicated geometrical computation, but its accuracy and efficiency are still unsatisfactory. An optimized volumetric visual hull reconstruction method is proposed based on CUDA. The voxel intersection judgment procedure and the isosurface extraction procedure are parallelized and implemented using multiple threads of CUDA to accelerate the shape modeling step. During triangulation step, marching cubes algorithm is improved through exact intersection points searching and redundant edge computation elimination. The pixel shader of CUDA is used to carry out the hybrid weight texture mapping so as to enhance the quality of the reconstructed mesh model. Experimental results on multi-camera platform DreamWorld proved the effectiveness of our method.

[1]  Tien-Tsin Wong,et al.  GPU-Friendly Marching Cubes for Visualizing Translucent Isosurfaces , 2007, MMVR.

[2]  Jean Ponce,et al.  On computing exact visual hulls of solids bounded by smooth surfaces , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[3]  Bruno Raffin,et al.  A Distributed Approach for Real Time 3D Modeling , 2004, 2004 Conference on Computer Vision and Pattern Recognition Workshop.

[4]  SkadronKevin,et al.  A performance study of general-purpose applications on graphics processors using CUDA , 2008 .

[5]  Chen Liang,et al.  Exact Visual Hull From Marching Cubes , 2008, VISAPP.

[6]  Hamish A. Carr,et al.  Accelerating marching cubes with graphics hardware , 2006, CASCON.

[7]  Wojciech Matusik,et al.  Polyhedral Visual Hulls for Real-Time Rendering , 2001, Rendering Techniques.

[8]  Sundaresan Raman,et al.  Quality Isosurface Mesh Generation Using an Extended Marching Cubes Lookup Table , 2008, Comput. Graph. Forum.

[9]  Yao Zhang,et al.  Parallel Computing Experiences with CUDA , 2008, IEEE Micro.

[10]  Thomas Malzbender,et al.  A Survey of Methods for Volumetric Scene Reconstruction from Photographs , 2001, VG.

[11]  Kunihiro Chihara,et al.  GPU-based shape from silhouettes , 2007, GRAPHITE '07.

[12]  Teresa C. S. Azevedo,et al.  3D Object Reconstruction from Uncalibrated Images using an Off-the-Shelf Camera , 2009 .

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

[14]  Junho Cho,et al.  Hardware-accelerated jaggy-free visual hulls with silhouette maps , 2006, VRST '06.

[15]  Edmond Boyer,et al.  Real-Time Capture, Reconstruction and Insertion into Virtual World of Human Actors , 2003, VVG.

[16]  David P. Luebke,et al.  CUDA: Scalable parallel programming for high-performance scientific computing , 2008, 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[17]  Kevin Skadron,et al.  A performance study of general-purpose applications on graphics processors using CUDA , 2008, J. Parallel Distributed Comput..

[18]  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.