公开(公告)号：US07139423B1

公开(公告)日：2006-11-21

申请号：US09831992

申请日：2000-09-14

申请人： Yannick Nicolas ,  Philippe Robert

发明人： Yannick Nicolas ,  Philippe Robert

IPC分类号： G06K9/00 ,  G06K9/36

CPC分类号： G06T7/579

摘要： The process comprises the following steps: calculating, for an image, a depth map corresponding to the depth, in the three-dimensional space, of image pixels; calculating, for an image, a resolution map corresponding to the three-dimensional resolution of the image pixels, based on the depth map; matching a current image pixel with a pixel of another image of the sequence, pixels relating to a common point in the three-dimensional scene, by projecting the current image pixel onto the other image; selecting a current image pixel according to its resolution and to that of the pixels of the sequence matched with this said pixel; constructing the three-dimensional model from the selected pixels. One application concerns the generation of images for browsing.

摘要翻译： 该处理包括以下步骤：针对图像计算与三维空间中的深度相对应的图像像素的深度图; 针对图像，基于深度图计算与图像像素的三维分辨率相对应的分辨率图; 通过将当前图像像素投影到另一图像上，将当前图像像素与该序列的另一图像的像素匹配，与三维场景中的公共点相关的像素; 根据其分辨率选择当前图像像素，并选择与该所述像素匹配的序列的像素的像素; 从所选择的像素构建三维模型。 一个应用程序涉及到用于浏览的图像的生成。

公开(公告)号：US07113635B2

公开(公告)日：2006-09-26

申请号：US10396182

申请日：2003-03-25

申请人： Philippe Robert ,  Yannick Nicolas

发明人： Philippe Robert ,  Yannick Nicolas

IPC分类号： G06K9/00 ,  G06T15/00

CPC分类号： G06T17/20

摘要： A process for modeling a 3D scene is provided which comprises validating the model by determining a maximum permitted distortion (5) on a 2D synthesized image (4), generated by the approximation inherent in the model (2); calculating for a point I of a reference image (7) and on a set of synthesized images (4) representing the 3D point of the scene corresponding to this point I, of the minimum (zi−Δzi1) and maximum (zi+Δzi2) depth values of the depth zi of this point I corresponding to this maximum distortion, calculating a span around the depth zi of this point I, dependant on the minimum value of the error Δzi2 and on the minimum value of the error Δzi1 among the values calculated for the synthesized images of the set.

摘要翻译： 提供了一种用于对3D场景进行建模的过程，其包括通过确定通过模型（2）中固有的近似生成的2D合成图像（4）上的最大允许失真（5）来验证模型。 对于参考图像（7）的点I和表示对应于该点I的场景的3D点的合成图像（4）的集合来计算最小（z 最大（z i i + 2）深度值 对应于该最大失真的这一点I的深度z i i i的深度z i根据误差的最小值计算围绕该点I的深度z i i的跨度 对于合成图像计算的值中的差值ΔLizi <0> 1 的最小值， 的集合。