Perspective (geometry)

About: Perspective (geometry) is a research topic. Over the lifetime, 277 publications have been published within this topic receiving 5795 citations. The topic is also known as: perspective (geometry).

3D environment labelling

Abstract: Objects (3, 4) of a three dimensional 3D virtual world that are represented on a two dimensional 2D image plane (6) are provided with labels (7, 8) which are generated in the 2D image plane (6) itself. The labels are in spatial registration with the object as represented on the 2D image plane. The generation of the labels in the 2D image plane avoids problems relating to perspective and aliasing which can occur when labels are provided in the virtual world itself and are rendered from the 3D virtual world.

Flexible calibration: minimal cases for auto-calibration

Abstract: This paper deals with the concept of auto-calibration, ie methods to calibrate a camera on-line In particular we deal with minimal conditions on the intrinsic parameters needed to make a Euclidean reconstruction, called flexible calibration The main theoretical results are that it is only needed to know that one intrinsic parameter is constant The method is based on an initial projective reconstruction, which is upgraded to a Euclidean one The number of images needed increases with the complexity of the constraints, but the number of points needed is only the number needed in order to obtain a projective reconstruction The theoretical results are exemplified in a number of experiments An algorithm, based on bundle adjustments and a linear initialization method are presented and experiments are performed on both synthetic and real data

External lead shape measurement apparatus for measuring lead shape of semiconductor package by using stereoscopic vision

Abstract: A lead shape measurement apparatus includes a first perspective imaging device for photographing external leads to be measured, from an oblique direction; a second perspective imaging device for photograping external leads to be measured, from an oblique direction different from that of the first perspective imaging device and so as to include a same photographic field of vision as the first perspective device; a first measurement point determination element for determining a center point of two intersecting virtual straight lines which intersect two ridges of the external leads to be measured in an image plane by photography of the first perspective imaging device; a second measurement point determination element for setting epipolar line in the image plane by the second perspective imaging device on the basis of sight lines when a measurement sampling point of the first perspective imaging device is photographed, and for determining a center point of two intersecting virtual straight lines which intersect two ridges of the external leads to be measured in an image plane and which exists upon the epipolar line; and a coordinate obtaining element for obtaining each of a plurality of the measurement sampling points which are set for measuring a three-dimensional shape of the external leads.

Euclidean structure from uncalibrated images

Abstract: A number of recent papers have demonstrated that camera “self-calibration” can be accomplished purely from image measurements, without requiring special calibration objects or known camera motion. We describe a method, based on self-calibration, for obtaining (scaled) Euclidean structure from multiple uncalibrated perspective images using only point matches between views. The method is in two stages. First, using an uncalibrated camera, structure is recovered up to an affine ambiguity from two views. Second, from one or more further views of this affine structure the camera intrinsic parameters are determined, and the structure ambiguity reduced to scaled Euclidean. The technique is independent of how the affine structure is obtained. We analyse its limitations and degeneracies. Results are given for images of real scenes. An application is described for active vision, where a Euclidean reconstruction is obtained during normal operation with an initially uncalibrated camera. Finally, it is demonstrated that Euclidean reconstruction can be obtained from a single perspective image of a repeated structure