Registration Techniques for Deformable Objects

TL;DR

This paper discusses methods for non-rigid registration of deforming objects, focusing on matching scans with partial overlaps, occlusions, and small deformations without prior templates or explicit features.

Contribution

It introduces techniques for deformable object registration that handle partial overlaps and occlusions without relying on prior shape models or feature correspondences.

Findings

Effective registration of deforming objects with partial data

Handling occlusions and non-rigid deformations

No need for prior templates or explicit features

Abstract

In general, the problem of non-rigid registration is about matching two different scans of a dynamic object taken at two different points in time. These scans can undergo both rigid motions and non-rigid deformations. Since new parts of the model may come into view and other parts get occluded in between two scans, the region of overlap is a subset of both scans. In the most general setting, no prior template shape is given and no markers or explicit feature point correspondences are available. So, this case is a partial matching problem that takes into account the assumption that consequent scans undergo small deformations while having a significant amount of overlapping area [28]. The problem which this thesis is addressing is mapping deforming objects and localizing cameras in the environment at the same time.