Adaptive Iterative Closest Keypoint

Authors

Johan Ekekrantz, Andrzej Pronobis, John Folkesson, Patric Jensfelt

Year

2013

Conference Paper

Algorithm

AICK

Source

European Conference on Mobile Robots

Identifiers

DOI 10.1109/ECMR.2013.6698824

Abstract¶

Finding accurate correspondences between overlapping 3D views is crucial for many robotic applications, from multi-view 3D object recognition to SLAM. This step, often referred to as view registration, plays a key role in determining the overall system performance. In this paper, we propose a fast and simple method for registering RGB-D data, building on the principle of the Iterative Closest Point (ICP) algorithm. In contrast to ICP, our method exploits both point position and visual appearance and is able to smoothly transition the weighting between them with an adaptive metric. This results in robust initial registration based on appearance and accurate final registration using 3D points. Using keypoint clustering we are able to utilize a non exhaustive search strategy, reducing runtime of the algorithm significantly. We show through an evaluation on an established benchmark that the method significantly outperforms current methods in both robustness and precision.

Summary¶

Introduces Adaptive Iterative Closest Keypoint, a lightweight RGB-D registration method that updates keypoint correspondences in an ICP-like loop while adapting which matches are trusted. The paper is useful for robotics perception because it targets fast registration of overlapping 3D views, improving robustness for object recognition and SLAM without requiring a heavy global optimization pipeline.