What does it look like? An artificial neural network model to predict the physical dense 3D appearance of a large-scale object

作者：

Highlights：

• A benchmark for image-based 3D reconstruction in a realistic condition is presented.

• The surface of a large-scale object at several twisting angles is densely localized.

• Both the descriptive statistics and qualitative analysis are reported.

• Prediction of bending level of an object without knowing the material’s properties.

• A remarkably low RMSE of 8 and a high R2 of 1 were yielded when tested on 211 data.

摘要

•A benchmark for image-based 3D reconstruction in a realistic condition is presented.•The surface of a large-scale object at several twisting angles is densely localized.•Both the descriptive statistics and qualitative analysis are reported.•Prediction of bending level of an object without knowing the material’s properties.•A remarkably low RMSE of 8 and a high R2 of 1 were yielded when tested on 211 data.

论文关键词：Artificial neural network,Robot arm,Surface reconstruction,Metal twisting,ArUco marker

论文评审过程：Received 27 May 2021, Revised 9 June 2022, Accepted 6 July 2022, Available online 13 July 2022, Version of Record 20 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118106