Publications

This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

Search for Publication


Year(s) from:  to 
Author:
Keywords (separated by spaces):

Frankenhorse: Automatic Completion of Articulating Objects from Image-based Reconstruction

Alex Mansfield, Nikolay Kobyshev, Hayko Riemenschneider, Will Chang, Luc Van Gool
BMVC
2014

Abstract

Manual 3D modelling can create clean complete models but takes time and expert knowledge. Image-based reconstructions of objects are easy to create, but are far from complete and clean. While small holes can be completed with a smoothness prior, large holes require a higher-level understanding of the object. We present the first method to complete large holes in articulating objects by reconstructing and aligning sets of objects of the same class, using the well-reconstructed parts in each model to complete holes in the others, resulting in a `Frankenhorse' completion. Our proposed method is fully automatic, and still is able to handle articulation, intra-class variation, holes and clutter present in the reconstructions. This is achieved through our novel segmentation and clutter removal processes as well as by the use of a robust method for piecewise-rigid registration of the models. We show that our method can fill large holes even when only a small set of models with high variability and low reconstruction quality is available.


Link to publisher's page
@InProceedings{eth_biwi_01132,
  author = {Alex Mansfield and Nikolay Kobyshev and Hayko Riemenschneider and Will Chang and Luc Van Gool},
  title = {Frankenhorse: Automatic Completion of Articulating Objects from Image-based Reconstruction},
  booktitle = {BMVC},
  year = {2014},
  keywords = {}
}