WO2013014689A1 - Procédé et appareil de texturation de modèles tridimensionnels à l'aide d'images bidimensionnelles - Google Patents

Procédé et appareil de texturation de modèles tridimensionnels à l'aide d'images bidimensionnelles Download PDF

Info

Publication number
WO2013014689A1
WO2013014689A1 PCT/IT2012/000225 IT2012000225W WO2013014689A1 WO 2013014689 A1 WO2013014689 A1 WO 2013014689A1 IT 2012000225 W IT2012000225 W IT 2012000225W WO 2013014689 A1 WO2013014689 A1 WO 2013014689A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
image
dimensional
points
pxyz
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/IT2012/000225
Other languages
English (en)
Inventor
Francesca UCCHEDDU
Anna Pelagotti
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Publication of WO2013014689A1 publication Critical patent/WO2013014689A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T15/00—Three-dimensional [3D] image rendering
    • G06T15/04—Texture mapping
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T15/00—Three-dimensional [3D] image rendering
    • G06T15/10—Geometric effects
    • G06T15/20—Perspective computation
    • G06T15/205—Image-based rendering
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three-dimensional [3D] modelling for computer graphics

Definitions

  • the invention relates to a method and an apparatus for texturing three-dimensional digital models with digital images of different origin, i.e., photographic,, radiographic, etc.
  • mapping of the surface of the model with a two- dimensional image is meant mapping of the surface of the model with a two- dimensional image.
  • the two-dimensional images may be images of a photographic type, but according to the purpose to be achieved may also be radiological images, thermographic images, etc.
  • Texturing basically consists of the combination of the points of internal and/or external surfaces of the 3D model with corresponding points of the image to be textured.
  • the texturing process entails mapping of information of colour "carried" by the photographic image onto ' the points of the 3D surface, or onto the elements (generally triangles) of a mesh representing the 3D surface.
  • the purpose of the present invention is to overcome the drawbacks of known technologies and to propose a method and an apparatus for texturing 3D digital models of an object with 2D digital images of the same object, which will achieve a high level of correspondence and will be implementable in an even completely automatic way.
  • the advantages achieved consist in the complete automation of the texturing process, irrespective of the knowledge of the optical sensors used for digital acquisition of the 3D model and of the type of 2D image to be textured, and in general when it is not is possible to acquire the image to be textured and the model simultaneously.
  • a further advantage is represented by the capacity of the method to establish a real and precise correspondence between the image to be textured and the 3D model.
  • Specific advantages are those deriving from the possibility of automation, with saving in time and reproducibility of the results. This is particularly useful in the case of maps of extensive surfaces, such as towns that can be textured with multiple views, having available images of small portions. List of the drawings
  • Figure 1 shows an example of point ' of observation for the creation of a depth map of a model, with respect to a reference plane
  • Figures 2-4 show an example of subsequent selection of observation points, according to the method of the invention.
  • Figure 5 shows the two-dimensional images of the depth maps obtained in the first iteration of the algorithm for selection of the observation points, and the respective value of resemblance with the image to be textured, expressed as value of maximum mutual information (MMI) ;
  • Figure 6 shows the difference image between the preferred depth map and the external image to be textured
  • Figure 7 shows the two-dimensional images of the depth maps obtained in a subsequent iteration of the algorithm for selection of the observation points, and the respective value of resemblance with the image to be textured, expressed as value of maximum mutual information (MMI) ;
  • Figure 8 shows the difference image between the preferred depth map and the external image to be textured in two successive iterations of the selection step
  • Figure 9 shows a model textured with the method according to the invention, where the blank parts are parts not covered by the external image used and hence not mapped;
  • Figure 10 shows a second example of 3D model and an external image to be textured on the model
  • Figure 11 shows in the ' first row the depth maps of the model of Figure 10, selected in three successive iterations according to the method and in the second row the difference images with the image to be textured;
  • Figure 12 shows the 3D model of Figure 10 textured with the method according to the invention starting from the external image chosen.
  • Described with reference to the drawings is a . method according to the invention for automatic texturing of a digital model M in three dimensions of an object S with external two-dimensional digital images I of the same object, even acquired independently of the creation of the model, for example photographic images, or radiological images, or images of another type.
  • an outer surface IF on which a finite number of observation points Oi of the model are established, preferably distributed in ah equidistant manner in order to cover the entire surface on a regular polyhedron (for example, an icosahedron) .
  • a two-dimensional image MP is created, obtained by projecting the points Pxyz of the surface of the model M in a plane Li constructed as a function of the observation point Oi, in such a way as to be able to identify the pixels of the image MP in direct correspondence with points Pxyz of the model M ( Figure 1), and in such a way that the pixels of the two-dimensional images MP of the model M contain information regarding the distance between the corresponding points Pxyz of the model M and the pre-selected plane of projection Li.
  • the outer surface is assumed as a regular polyhedron constituted by faces of which the observation points represent the central point.
  • the selected observation point Osel will be the central point of the face of the polyhedron to which the selected image MPsel corresponds ( Figure 2).
  • the procedure can be re-iterated until a preferred two- dimensional image ⁇ (MPpref) is determined that meets a pre- set criterion of acceptance of resemblance with the external image I.
  • the points of the image MPpref have by construction a point-by-point correspondence with the surface of the model and have at the same time a strict correlation with the digital input image I, which increases as the number of automatic iterations increases ( Figures 7-8) until the pre-set criterion of acceptability is met.
  • the two-dimensional images of the model M. are constructed as depth maps, where the level of grey of the pixels of the image represents the distance of the corresponding point Pxyz of the investigated surface of the model M from the plane Li of projection.
  • the depth maps are constructed starting from the model M, as a two-dimensional array, so that the co-ordinates x, y of the points of the model correspond to the rows and the columns of the matrix (as in an ordinary image) , whilst the corresponding values of depth (value z of the points of the model) refer to a plane of section of the model (or an external plane) and are stored in the value of the element of the array.
  • Said map is then constructed as a grey-scale image in which the information z is stored in the intensity-of-grey information.
  • the images MP are hence a projection of appropriate surfaces of the model M in the two-dimensional plane Li.
  • the maps MP constructed in this way thus intrinsically maintain an exact correspondence between the 2D pixels and the 3D points of the model.
  • the maps MP constructed in this way thus intrinsically maintain an exact correspondence between the 2D pixels and the 3D points of the model.
  • step of selection of the "best" image MPpref can be executed automatically, for example, as has been said, by entering a predefined value of measurement of resemblance, such as the mutual information itself or other recording methods .
  • the method is suited for automatic execution via a computer program appropriately instructed with the initial pre-selected data, i.e., the model M to be textured, the image or images I of the subject S, acquired even at different times and/or with different techniques, and an initial ensemble of preferences (number and distribution of the observation points Oi, choice of the plane of projection Li, number of iterations or criteria of resemblance threshold in recording the 2D-2D images.
  • the initial pre-selected data i.e., the model M to be textured
  • the image or images I of the subject S acquired even at different times and/or with different techniques
  • an initial ensemble of preferences number and distribution of the observation points Oi, choice of the plane of projection Li, number of iterations or criteria of resemblance threshold in recording the 2D-2D images.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Graphics (AREA)
  • General Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Image Analysis (AREA)
  • Woven Fabrics (AREA)
  • Looms (AREA)

Abstract

L'invention concerne un procédé de texturation d'un modèle numérique tridimensionnel (M) d'un objet, constitué de points dans l'espace (Pxyz) avec une image d'entrée numérique bidimensionnelle (I), lequel procédé comprend les étapes consistant à construire des images bidimensionnelles (MPi) obtenues par projection des points (Pxyz) du modèle (M) dans un plan (Li), à établir une relation de ressemblance maximale entre l'image d'entrée avec l'une des images construites (image tridimensionnelle (3D)), et à réitérer lesdites étapes jusqu'à ce qu'une image bidimensionnelle construite du modèle soit déterminée dans une correspondance point-pixel directe avec l'image d'entrée numérique (I).
PCT/IT2012/000225 2011-07-22 2012-07-19 Procédé et appareil de texturation de modèles tridimensionnels à l'aide d'images bidimensionnelles Ceased WO2013014689A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IT000150A ITFI20110150A1 (it) 2011-07-22 2011-07-22 Metodo e apparato per la tessitura di modelli tridimensionali con immagini bidimensionali
ITFI2011A000150 2011-07-22

Publications (1)

Publication Number Publication Date
WO2013014689A1 true WO2013014689A1 (fr) 2013-01-31

Family

ID=44675667

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IT2012/000225 Ceased WO2013014689A1 (fr) 2011-07-22 2012-07-19 Procédé et appareil de texturation de modèles tridimensionnels à l'aide d'images bidimensionnelles

Country Status (2)

Country Link
IT (1) ITFI20110150A1 (fr)
WO (1) WO2013014689A1 (fr)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2006703A2 (fr) * 2007-06-21 2008-12-24 F.Hoffmann-La Roche Ag Systèmes et procédés d'alignement d'objets dans des images

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2006703A2 (fr) * 2007-06-21 2008-12-24 F.Hoffmann-La Roche Ag Systèmes et procédés d'alignement d'objets dans des images

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
IOANNIS STAMOS ET AL: "Integrating Automated Range Registration with Multiview Geometry for the Photorealistic Modeling of Large-Scale Scenes", INTERNATIONAL JOURNAL OF COMPUTER VISION, KLUWER ACADEMIC PUBLISHERS, BO, vol. 78, no. 2-3, 10 November 2007 (2007-11-10), pages 237 - 260, XP019581884, ISSN: 1573-1405 *
REMONDINO F ET AL: "Novel data registration techniques for art diagnostics and 3D heritage visualization", PROC. OF IX CONF. ON OPTICAL 3-D MEASUREMENT TECHNIQUES, vol. 1, 1 July 2009 (2009-07-01) - 3 July 2009 (2009-07-03), Vienna, Austria, pages 1 - 10, XP008150287 *

Also Published As

Publication number Publication date
ITFI20110150A1 (it) 2013-01-23

Similar Documents

Publication Publication Date Title
CN111598993B (zh) 基于多视角成像技术的三维数据重建方法、装置
CN114329747B (zh) 一种面向建筑数字孪生的虚实实体坐标映射方法及系统
CN113689578B (zh) 一种人体数据集生成方法及装置
US12033276B2 (en) Learning-based 3D model creation apparatus and method
CN111640180B (zh) 一种三维重建方法、装置及终端设备
US9965870B2 (en) Camera calibration method using a calibration target
JP7005622B2 (ja) 認識処理装置、認識処理方法及びプログラム
EP3622481B1 (fr) Procédé et système d'étalonnage de système de vélocimétrie
CN107977992A (zh) 一种基于无人机激光雷达的建筑物变化检测方法及装置
CN113052907A (zh) 一种动态环境移动机器人的定位方法
CN109559349A (zh) 一种用于标定的方法和装置
CN106705849A (zh) 线结构光传感器标定方法
CN113781538B (zh) 一种图像深度信息融合方法、系统、电子设备及存储介质
CN113724365B (zh) 一种三维重建方法及装置
JP2024064413A (ja) 画像データ生成装置、画像データ生成方法、画像データ生成プログラム
CN116469101A (zh) 数据标注方法、装置、电子设备及存储介质
JP6306903B2 (ja) 情報処理装置、情報処理装置の制御方法およびプログラム
WO2019116541A1 (fr) Dispositif de traitement d'informations et procédé de simulation de région mesurable
WO2019150431A1 (fr) Dispositif de traitement d'informations
KR101725166B1 (ko) 2차원 이미지들을 이용하여 3차원 이미지를 재구성하는 방법 및 이를 위한 장치
WO2013014689A1 (fr) Procédé et appareil de texturation de modèles tridimensionnels à l'aide d'images bidimensionnelles
CN109829951B (zh) 平行等位检测方法、装置与自动驾驶系统
CN117456493A (zh) 目标检测方法、装置、计算机设备和存储介质
CA2956492A1 (fr) Procede, dispositif mobile et programme informatique permettant de substituer une surface de revetement d'ameublement dans une image
JP2002135807A (ja) ３次元入力のためのキャリブレーション方法および装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12753841

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12753841

Country of ref document: EP

Kind code of ref document: A1