CoBigICP: コレントロピー指標と双方向対応を用いたロバストかつ高精度な点群位置合わせ
CoBigICP: Robust and Precise Point Set Registration using Correntropy Metrics and Bidirectional Correspondence
局所的な幾何情報と大域的なノイズ特性を組み合わせ、外れ値に強いコレントロピー誤差指標と双方向対応を導入した新しいICP変種を提案し、既存手法を上回る精度を示した。
著者: Pengyu Yin, Di Wang, Shaoyi Du, Shihui Ying, Yue Gao, Nanning Zheng
分類: cs.RO
原文アブストラクト
In this paper, we propose a novel probabilistic variant of iterative closest point (ICP) dubbed as CoBigICP. The method leverages both local geometrical information and global noise characteristics. Locally, the 3D structure of both target and source clouds are incorporated into the objective function through bidirectional correspondence. Globally, error metric of correntropy is introduced as noise model to resist outliers. Importantly, the close resemblance between normal-distributions transform (NDT) and correntropy is revealed. To ease the minimization step, an on-manifold parameterization of the special Euclidean group is proposed. Extensive experiments validate that CoBigICP outperforms several well-known and state-of-the-art methods.
