Tele-Registration: A Field-Guided Registration for Feature-Conforming Shape Composition

ACM Transactions on Graphics 2012

(Proceedings of SIGGRAPH Asia 2012)

Hui Huang1   Minglun Gong2  Daniel Cohen-Or3   Yaobin Ouyang1   Fuwen Tan1   Hao Zhang4 

  1VisuCA/SIAT   2Memorial University    3Tel Aviv University   4Simon Fraser University

Figure 1: Composing parts, possibly with sharp features and non-overlapping boundaries, presents challenges to both part alignment and blending. Our field-guided approach (see middle for a visualization of the fields) leads to alignment of parts away from each other and feature-conforming surface blending. The bridging surfaces generated (colored yellow on the right) are piecewise smooth. 

We present an automatic shape composition method to fuse two shape parts which may not overlap and possibly contain sharp features, a scenario often encountered when modeling man-made objects. At the core of our method is a novel field-guided approach to automatically align two input parts in a feature-conforming manner. The key to our field-guided shape registration is a natural continuation of one part into the ambient field as a means to introduce an overlap with the distant part, which then allows a surface-to-field registration. The ambient vector field we compute is feature-conforming; it characterizes a piecewise smooth field which respects and naturally extrapolates the surface features. Once the two parts are aligned, gap filling is carried out by spline interpolation between matching feature curves followed by piecewise smooth least-squares surface reconstruction. We apply our algorithm to obtain feature-conforming shape composition on a variety of models and demonstrate generality of the method with results on parts with or without overlap and with or without salient features. 

Figure 2: Transplanting of shape parts: broken perfume bottles lying on the floor are aligned and blended with the same bottom part shown in green. All parts used in these examples have non-flat boundaries. The top and bottom parts to be composed may differ in size, shape, and boundary and feature characteristics, necessitating field-guided registration and blending with non-conforming feature profiles.

Figure 3: Composition of parts with more than one open boundaries. The parts come from the models on the left; in the middle we present automatic field-guided registration results, where the alignment step searches for a rigid transformation based on multiple pairs of open boundaries; the right side shows two final models obtained after gap filling.

Figure 4: Handling of partial overlap. The optimal alignment found by our registration overlaps the two parts; see zoom-ins. In these cases, the bridging surfaces are constructed by applying an automatic backward mesh receding step before gap filling. Note that one foot (marked by black boxes) is facing the wrong direction since the registration is based on geometric match around the open boundaries.

Figure 5: Composition with unmatched feature counts. The chair has four-sided open boundaries, whereas the legs of the green, blue, and purple robots have three-, six-, and eight-sided ones, respectively. The zoom-ins reveal how the bridging surfaces connect different boundaries and blend feature curves (highlighted with yellow) into smooth regions. Note however that our field-guided registration still works effectively.

Figure 6: Robustness to noise in the composed parts. Pre-smoothing via bilateral filtering is sufficiently robust for our registration to behave well.


Figure 7: Centaur model obtained with and without user specified constraints.
The authors would like to thank all the reviewers for their valuable comments. Thanks also go to Mark Schmidt for his Matlab code minFunc, which has been modified to solve our unconstrained optimization problem. The models in Figures 11 and 13 are courtesy of the AIM@SHAPE Shape Repository. This work is supported in part by grants from NSFC (61103166 and 61232011), Guangdong Science and Technology Program (2011B050200007), National 863 Program (2011AA010503), Shenzhen Science and Innovation Program (CXB201104220029A), Natural Science and Engineering Research Council of Canada (293127 and 611370) and the Israel Science Foundation.

title = {Field-Guided Registration for Feature-Conforming Shape Composition},
author = {H. Huang and M. Gong and D. Cohen-Or and Y. Ouyang and F. Tan and H. Zhang},
journal = {ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia 2012)},
volume = {31},
issue = {6},
pages = {171:1-171:11},
year = {2012},
Copyright © 2016-2018 Visual Computing Research Center