| 摘要 |
3D reconstruction of at least part of a first real object at absolute spatial scale comprises: receiving a first and third image including at least part of the first real object captured with a first and third camera, a second and fourth image including at least a first and second part of a second real object, captured with a second and fourth camera, wherein the frustum of the first and second as well as the third and fourth camera do not overlap, and providing a first and second spatial transformation between the first and second and the third and fourth camera, providing a first and second scale information indicative of an absolute spatial scale of the at least first and second part of the second real object, determining at least part of a pose of the second (P2F) and fourth (P4F) camera in a second common coordinate system according to the second and fourth image and the first and second scale information, determining at least part of a pose of the first (P1F) and third (P3F) camera according to the at least part of the pose of the second (P2F) and fourth (P4F) camera and the first and second spatial transformation, and determining said 3D reconstruction according to the first and third images and the at least part of the poses of the first (P1F) and third (P3F) cameras. |
