Pursuing 3-D Scene Structures With Optical Satellite Images From Affine Reconstruction to Euclidean Reconstruction

How to use multiple optical satellite images to recover the 3-D scene structure is a challenging and important problem in the remote sensing field. Most existing methods in literature have been explored based on the classical rational polynomial coefficients (RPCs) camera model which requires at least 39 ground control points (GCPs), however, it is nontrivial to obtain such a large number of GCPs in many real scenes. Addressing this problem, we propose a hierarchical reconstruction framework based on multiple optical satellite images, which needs only four GCPs to fully-automated reconstruct the 3-D scene structure. The proposed framework is independent of the RPC model and composed of a dense affine reconstruction stage and a followed affine-to-Euclidean upgrading stage: At the dense affine reconstruction stage, a dense affine reconstruction approach is explored for pursuing the 3-D affine scene structure without any GCP from input satellite images. Then at the affine-to-Euclidean upgrading stage, the obtained 3-D affine structure is upgraded to a Euclidean one with four GCPs. Experimental results on two public datasets demonstrate that the proposed method significantly outperforms several state-of-the-art methods in most cases.

Automated summary

In this study, a hierarchical reconstruction framework based on multiple optical satellite images is proposed to recover the 3-D scene structure. With only four ground control points (GCPs), the proposed framework achieves fully automated reconstruction and outperforms several state-of-the-art methods in most cases.