“Iterative Cage-Based Registration for Dynamic Shape Capture” by Savoye

  • ©Yann Savoye

  • ©Yann Savoye

  • ©Yann Savoye




    Iterative Cage-Based Registration for Dynamic Shape Capture



    Recent advances in low-cost dynamic scanning turn the cross-parametrization of non-rigid animatable surface into a vision-oriented ill-posed problem. In contrast with [Li et al. 2012], we propose a novel detail-preserving registration approach with resolution-independent control. Furthermore, our skin-detached surface registration avoids patch-based segmentation or affine fitting to maintain the local plasticity, as required in [Budd and Hilton 2010]. In particular, we leverage the problem of highly non-rigid spacetime registration by employing an elasto-plastic coarse cage. Thus, we perform scalable handle-aware harmonic shape registration, relying on the high-level of shape abstraction offered by the space-based paradigm. To the best of our knowledge, our technique is the first to investigate handle-aware elastic overlapping-rigidities for registering life-like dynamic shapes in full-body clothing.


    1. Budd, C., and Hilton, A. 2010. Temporal alignment of 3d video sequences using shape and appearance. In CVMP’10: 9th Conference on Visual Media Production.
    2. Jacobson, A., Baran, I., Popović, J., and Sorkine, O. 2011. Bounded biharmonic weights for real-time deformation. ACM Trans. Graph. 30.
    3. Li, H., Luo, L., Vlasic, D., Peers, P., Popović, J., Pauly, M., and Rusinkiewicz, S. 2012. Temporally coherent completion of dynamic shapes. ACM Trans. Graph. 31.
    4. Vlasic, D., Baran, I., Matusik, W., and Popović, J. 2008. Articulated mesh animation from multi-view silhouettes. ACM Trans. Graph. 27 (August), 97:1–97:9.

Additional Images:

©Yann Savoye

ACM Digital Library Publication:

Overview Page: