Efficient and Handy Texture Mapping on 3D Surfaces

Abstract

There has been a rapid technical progress in three‐dimensional (3D) computer graphics. But gathering surface and texture data is yet a laborious task. This paper addresses the problem of mapping photographic images on the surface of a 3D object whose geometric data are already known. We propose an efficient and handy method for acquiring textures and mapping them precisely on the surface, employing a digital camera alone. We describe an algorithm for selecting a minimal number of camera positions that can cover the entire surface of a given object and also an algorithm to determine camera’s position and direction for each photograph taken so as to paste it to the corresponding surfaces precisely. We obtained a matching accuracy within a pixel on a surface through three experimental examples, by which the practicability of our method is demonstrated.

Robust 2D-3D alignment based on geometrical consistency

“…Zuffi [17] applies this algorithm for the pose estimation of a knee joint in a single X-ray image for total knee replacement surgery. In the contour-based approach, the error is computed as the sum of distances between points on a contour line in a 2D image and on a projected contour line of a 3D model [2], [13], [14]. Iwashita [5] proposed the fast alignment algorithm utilizing a 2D distance map constructed by Level Set Method.…”

A Silhouette-Based Algorithm for Texture Registration and Stitching

“…In the field of capturing surface appearance (color and texture) of real world objects there have been a number of recent publications ranging from architectural scenes [5,23] to smaller artifacts [11,12,14,18,19] and even deformable objects like faces [7,8,15]. To acquire a complete texture for an object the following tasks must be performed.…”

“…A set of camera positions must be determined from which every part of the surface is captured by at least one image. For a given geometric model and a set of possible positions Matsushita and Kaneko [12] determine the optimal set of required views respecting the viewing angle. Further, Stuerzlinger [21] finds a minimal set of view points within the volume of all possible camera positions.…”

“…A lot of previous algorithms try to find the camera transformation by minimizing the error between the contour found in the image and the contour of the projected 3D model [4,8,10,12,14]. The error is typically computed as the sum of distances between a number of sample points on one contour to the nearest points on the other [12,14].…”

“…The error is typically computed as the sum of distances between a number of sample points on one contour to the nearest points on the other [12,14]. Another approach computes the sum of minimal distances of rays from the eye point through the image contour to the model's surface which is computed using 3D distance maps [4].…”