TY - GEN

T1 - Registration of 3D triangular models to 2D X-ray projections using black-box optimisation and X-ray simulation

AU - Wen, Tianci

AU - Mihail, Radu P.

AU - Al-Maliki, Shatha

AU - Letang, J.M.

AU - Vidal, Franck

PY - 2019/9/12

Y1 - 2019/9/12

N2 - Image registration has been studied extensively for the past few decades. In this paper we propose a novel idea to solve the registration of 3D triangular models onto 2D projections. Our approach relies extensively on global optimisation methods and fast X-ray simulation on GPU. We demonstrate the validity of our approach on two registration problems: i) 3D kinematic configuration of a 3D hand model, i.e. the recovery of the original hand pose from a postero-anterior (PA) view radiograph. ii) Automatic estimation of the position and rigid transformation of geometric shapes (cube and cylinders) to match an actual metallic sample made of Ti/SiC fibre composite with tungsten (W) cores. The performance is measured by Mean Absolute Error (MAE) and Zero-mean Normalised Cross Correlation (ZNCC). To evaluate our pipeline, each optimisation is repeated 15 times to gather statistically meaningful results, in particular to assess the reproducibility of the outputs. Our registration framework is successful for both test-cases when using a suitable optimisation algorithm.

AB - Image registration has been studied extensively for the past few decades. In this paper we propose a novel idea to solve the registration of 3D triangular models onto 2D projections. Our approach relies extensively on global optimisation methods and fast X-ray simulation on GPU. We demonstrate the validity of our approach on two registration problems: i) 3D kinematic configuration of a 3D hand model, i.e. the recovery of the original hand pose from a postero-anterior (PA) view radiograph. ii) Automatic estimation of the position and rigid transformation of geometric shapes (cube and cylinders) to match an actual metallic sample made of Ti/SiC fibre composite with tungsten (W) cores. The performance is measured by Mean Absolute Error (MAE) and Zero-mean Normalised Cross Correlation (ZNCC). To evaluate our pipeline, each optimisation is repeated 15 times to gather statistically meaningful results, in particular to assess the reproducibility of the outputs. Our registration framework is successful for both test-cases when using a suitable optimisation algorithm.

U2 - 10.2312/cgvc.20191265

DO - 10.2312/cgvc.20191265

M3 - Conference contribution

SN - 978-3-03868-096-3

SP - 105

EP - 113

BT - Computer Graphics & Visual Computing (CGVC)

PB - The Eurographics Association

T2 - Annual Conference in Computer Graphics & Visual Computing (CGVC)

Y2 - 12 September 2019

ER -