• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to a method of compressing data representative of a digital three-dimensional object, comprising: - the acquisition (60) of a three-dimensional object (2), comprising: • a mesh (4) formed of a plurality of polygons; • an atlas of texture (6) of the polygons of the mesh; - the simplification (70) of the mesh, by removing vertices, which removes polygons and instead creates new polygons with a larger face; and wherein, before deleting vertices, the method includes: identifying (64), in the mesh, first (12) and second (14) adjacent polygons that have different textures, and then: first and second vertices in common by creating third (B ') and fourth (C') vertices.
    公开号:FR3028990A1
    申请号:FR1461318
    申请日:2014-11-21
    公开日:2016-05-27
    发明作者:Florent Dupont；Guillaume Lavoue；Laurent Chevalier
    申请人:LYON 2 A ET L LUMIERE, University of；Centre National de la Recherche Scientifique CNRS；Universite Claude Bernard Lyon 1 UCBL；Ecole Centrale de Lyon；Institut National des Sciences Appliquees de Lyon ；
    IPC主号:
    专利说明:

    [0001] METHODS FOR COMPRESSING AND DECOMPRESSING DATA REPRESENTATIVE OF A THREE-DIMENSIONAL DIGITAL OBJECT AND INFORMATION RECORDING MEDIUM CONTAINING THE SAME [1] The invention relates to a method of compressing data representative of a three-dimensional object. The invention further relates to a set of data representative of a compressed three-dimensional object obtained by means of this compression method, as well as an information recording medium containing this set of data. The invention also relates to a method of decompressing such data that has been compressed using this method. The invention finally relates to an information recording medium and a computer for implementing these methods. [2] Typically, in the field of computer vision and three-dimensional computer graphics ("computer graphics" or "computer vision" in English), a three-dimensional object is represented numerically in the form of a mesh ("polygonal"). mesh "in English). This mesh is formed of a plurality of planar polygons contiguous with each other. Each polygon has a plurality of vertices interconnected by edges that delimit one face of the polygon. [3] "Progressive" compression processes in English are known. These methods facilitate in particular the transmission of a three-dimensional object from a multimedia content server to a terminal of a client on which the object is to be displayed. In these methods, the mesh is gradually simplified by operations of selective deletion (operations also called "decimation") of the vertices, to reduce their size. The simplified mesh is transmitted to the terminal, where it is displayed. Then it is gradually reconstructed on this terminal from incremental data transmitted thereafter, until the three-dimensional object as it was initially before compression. [4] An example of this method is described in the document: "Rate-distortion optimization for progressive compression of 3D meshes with color attributes", Ho Lee et al, The Visual Computer, vol. 28, p. 137-153, Springer-Verlag, May 2011, DOI: 10. 1007 / s00371 -011-0602-y. During the progressive display of the three-dimensional object being decompressed, these methods have the disadvantage of generating graphic artifacts when used with textured three-dimensional objects. By textured object, we mean a three-dimensional object in which the polygons of the mesh have their surface covered by a digital image, called "texture" ("texture mapping" in English). 3028990 2 [6] There is therefore a need for a method of progressive compression of a three-dimensional object that limits the appearance of such graphic artifacts on the object when it includes textures. [7] The invention therefore relates to a method for compressing data representative of a digital three-dimensional object, comprising: the acquisition of data representative of a three-dimensional object, these data comprising: a mesh formed of a plurality planar polygons contiguous with each other, each polygon comprising: a plurality of vertices interconnected by edges which delimit one face of the polygon, and a texture covering the face of the polygon; - an atlas of texture in which are included all the textures of the polygons of the mesh; - the simplification of the mesh, by removing vertices from the mesh, which eliminates polygons and instead creates new polygons with a larger face, including: - the identification, according to a predetermined criterion, of vertices to delete, - the deletion of the vertices identified and the edges connecting these vertices to 20 other vertices of the mesh, so as to eliminate these polygons comprising these vertices and these edges, - the creation: - new edges to connect the vertices which have not been deleted; new textures for these new polygons from the respective textures of the deleted polygons; and wherein, before deleting vertices, the method comprises: - identifying, in the mesh, first and second adjacent polygons that have different textures, two polygons being said to be adjacent if they present a first and a second vertex in common interconnected by a common edge, then - the replacement in the second polygon of the first and second vertex in common by, respectively, third and fourth vertices, these third and fourth vertices occupying the same positions in space as respectively, the first and second vertices; O the creation of two edges of zero length between, respectively, the first and third vertices and the second and fourth vertices and therefore at least one intermediate polygon of zero surface interposed between the first and second polygons. [8] The inventors have identified that a problem arises when the three-dimensional object has polygons adjacent to one another and whose textures are different. The problem arises particularly at the level of boundaries between portions of the object that have textures very different from each other.
    [0002] Typically, the known methods of progressive compression select the vertices to be suppressed as a function of the geometric properties of the mesh, without taking into account the texture information of the polygons.  Thus, the common edge separating these adjacent polygons can be suppressed due to simplification.  The information concerning the fact that, before simplification of the mesh, there existed different textures on each side of the removed edge is thus absent in the simplified mesh obtained.  Therefore, when decompressing, when this simplified mesh is displayed, a graphical artifact appears.  This degrades the visual quality of the three-dimensional object perceived during its decompression.  [9] In the above method, the search of the first and second polygons 15 makes it possible to identify the polygons where an artifact would be likely to appear if the simplification was directly implemented.  The replacement in the second polygon of the first and second vertex in common by the third and fourth distinct vertices and the creation of the edges make it possible to separate the first and second polygons.  They are no longer adjacent to each other.  Thus, for example, the deletion of the first or second vertex does not result in the disappearance of both the first and second polygons.  During compression, the texture difference information is thus preserved at least until the first, second, third and fourth vertices have been removed from the simplified mesh.  This information is therefore preserved longer during the compression process.  Therefore, conversely, during decompression, this information reappears much earlier in the simplified mesh displayed on the screen.  This minimizes the time during which such an artifact is visible which improves the visual quality of the object displayed during decompression.  Finally, since the third and fourth vertices occupy the same position as the first and second vertices and the length of the edges is zero, the external appearance of the object is not modified by this creation of additional vertices. .  Embodiments of the invention may have the following characteristic: the method comprises, before deleting vertices, the marking of the first, second, third and fourth vertices; and during the simplification of the mesh, the first, second, third and fourth vertices which have been the subject of said marking are not identified as vertices to be deleted even if they satisfy the predetermined criterion.  According to another aspect, the invention also relates to a set of data representative of a compressed three-dimensional digital object, these data comprising: a simplified mesh formed of a plurality of polygons and obtained by means of a compression method according to the invention; incremental decompression data, these data including a list of vertices and edges that have been removed from the mesh when performing said compression method; and wherein at least one of the simplified mesh and incremental decompression data encodes first, second, third and fourth vertexes of the mesh such that the first and third vertices are interconnected by an edge of zero length and the second and fourth vertices are interconnected by an edge of zero length.  In another aspect, the invention also relates to an information recording medium, comprising the set of data representative of a compressed digital three-dimensional object according to the invention.  According to another aspect, the invention also relates to a method of decompressing data representative of a compressed three-dimensional digital object, comprising: the acquisition of a set of data representative of a compressed three-dimensional digital object to the invention; the reconstruction of a more complex mesh from the simplified mesh and the incremental data, this reconstruction, comprising the creation of: additional vertices in the simplified mesh from the acquired incremental data and the replacement of edges of the simplified mesh by additional edges connecting these new vertices to existing vertices, so as to remove polygons from the simplified mesh and replace them with additional polygons having a smaller area; additional textures, for these additional polygons, from the respective textures of the replaced polygons and incremental data, and in which the more complex reconstructed mesh comprises the first and third vertices interconnected by an edge of zero length and the second and third vertices fourth vertices interconnected by an edge of zero length.  Embodiments of the invention may have the following characteristic: the reconstructed mesh is shaped by replacing the third and fourth vertices of the second polygon by, respectively, the first and second vertices and, by removing the edges of zero length between the first and third vertices on the one hand and the second and fourth vertices on the other hand.  According to another aspect, the invention also relates to a method for transmitting data representative of a digital three-dimensional object between a transmitter and a receiver, this method comprising: - acquisition of representative data of an object digital by the issuer; compression of the acquired data by means of a compression method according to the invention; the transfer of the compressed data from the transmitter to the receiver by means of a data exchange link; the decompression, by the receiver, of the compressed data transferred, by means of a decompression method according to the invention.  Embodiments of the invention may have one or more of the following features: two textures are said to be different if their intersection in the atlas of textures is equal to the null set.  two textures are said to be different if the shortest distance d separating these two textures in the texture atlas is strictly greater than a predetermined threshold.  According to another aspect, the invention also relates to an information recording medium, comprising instructions for the execution of a method according to the invention when these instructions are executed by an electronic computer.  In another aspect, the invention finally relates to an electronic computer 20 for implementing the invention, the electronic computer being programmed to: - acquire data representative of a three-dimensional object, these data comprising: - a mesh formed of a plurality of planar polygons contiguous with each other, each polygon comprising: a plurality of vertices interconnected by edges which delimit one face of the polygon, and a texture covering the face of the polygon; an atlas of texture in which all the textures of the polygons of the mesh are included; simplifying the mesh, by removing vertices from the mesh, which eliminates polygons and instead creates new polygons having a larger face, by: identifying, according to a predetermined criterion, vertices to be removed, - removing vertices identified and edges connecting these vertices to other vertices of the mesh, so as to remove these polygons with these vertices and edges, - creating: 3028990 6 - new edges to connect vertices that have not been removed; - new textures, for these new polygons, from the respective textures of the deleted polygons; the computer being programmed to, before deleting vertices,: identifying, in the mesh, first and second adjacent polygons that have different textures, two polygons being said to be adjacent if they present a first and a second vertex in common interconnected by a common edge, then: - replacing in the second polygon first and second vertex in common by, respectively, third and fourth vertices, these third and fourth vertices occupying the same positions in space as, respectively first and second vertices; creating two zero-length edges between, respectively, the first and third vertices and the second and fourth vertices and thus creating at least one zero-surface intermediate polygon interposed between the first and second polygons.  The invention will be better understood on reading the description which follows, given solely by way of nonlimiting example and with reference to the drawings, in which: FIG. 1 is a diagrammatic illustration of a three-dimensional digital object; FIG. 2 is a schematic illustration of a set of data representative of a portion of the object of FIG. 1; Figure 3 is an illustration of a device for transmitting the data of Figure 2; FIG. 4 is a flow chart of a data compression method of FIG. 2; FIG. 5 is a schematic illustration of the portion of the object of FIG. 2 as modified at the end of a step of the method of FIG. 4; FIGS. 6A to 6C are schematic illustrations of portions of the mesh of the object of FIG. 1 as modified during the execution of a step of the method of FIG. 4; FIGS. 7A and 7B are schematic illustrations of the object of FIG. 1 after application of the compression method of FIG. 4; FIGS. 8A and 8B are schematic illustrations of the object of FIG. 1 after application of the compression method of FIG. 4 when a vertex duplication step is omitted from this method; FIG. 9 is a schematic illustration of an information recording medium which comprises a set of data representative of the three-dimensional digital object compressed at the end of the process of FIG. 4; FIG. 10 is a flow chart of a method of decompressing the objects of FIGS. 7A-8B to reconstruct the object of FIG. 1; FIG. 11 is a schematic illustration of another embodiment of a step of the method of FIG. 4.  In these figures, the same references are used to designate the same elements.  [0022] In the remainder of this description, the features and functions well known to those skilled in the art are not described in detail.  Figure 1 shows a three-dimensional digital object 2 on which is applied a texture in the form of a checkerboard.  This object 2 comprises a mesh 4 and an atlas 6 of textures.  Mesh 4 ("polygonal mesh" in English) is formed of a plurality of planar polygons contiguous to each other.  Each polygon includes: - vertices ("vertex" in English), - edges that connect the vertices of the polygon two to two, 20 - a face, delimited by the edges.  The mesh 4 does not have any holes here.  The atlas 6 comprises a plurality of individual textures, each of which covers a face of a polygon of the mesh 4.  Each texture is a two-dimensional image, for example a matrix image.  This technique is known as "texture mapping" in the English language.  The polygons here are triangular in shape.  Indeed, most graphical processing units in English are optimized for the processing of triangular polygons.  Also, in what follows, we can use the term triangle to designate these polygons.  In this example, object 2 schematically represents a frog.  The atlas 6 has a checkerboard pattern of black and white color, which covers the entire outer surface of the mesh 4.  FIG. 2 represents a set of digital data used to represent the object 2.  For simplicity, these data are described for only two triangles 12 and 14 of the mesh 4.  These two triangles 12, 14 are also represented in an enlarged version in FIG.  In the following, the term object 2 is used to designate data representative of object 2.  We will speak of acquisition of the object 2 to designate the acquisition of the data 10.  The assembly 10 comprises: 40 - atlas 6 of textures; 3028990 8 - a list 22 of the vertices of the polygons of the mesh 4; a list 24 of the polygons of the mesh 4; a list 26 of texture coordinates of the polygons of the mesh 4.  The atlas 6 contains all the textures necessary to fill each face of each polygon of the mesh 4.  Each texture of a face of a polygon is indexed within atlas 6 by a set of three coordinates (Ua, Va), (Ub, Vb) and (Us, Vs) ("UV mapping" in English). ).  For example, the atlas 6 contains a two-dimensional matrix image 27.  The coordinates (Ua, Va), (Ub, Vb) and (Us, Vs) encode the position of three points in an orthonormal frame of this image 27.  These three points 10 delimit a piece of the image 27 also called "piece of texture".  It is this piece of texture that is used to fill the face of a polygon.  In what follows, these coordinates will be referred to as "UV coordinates".  Specifically, here, each piece of texture has the same shape as the polygon to which it corresponds.  Thus, each piece of texture has a triangular shape that has exactly the same dimensions as the face of the polygon to be filled with that piece of texture.  However, the vertices of the texture pieces are not vertices of the mesh 4.  To avoid confusion, we speak of "point" to designate the vertices of the pieces of texture.  The texture pieces corresponding to the polygons 12 and 14 respectively carry the references 28 and 29 in the image 27.  We note their respective texture coordinates (Uia, Via), (Uib, Vib) and (U1s, V1,) for the piece 28 and (U2a, V2a), (U2b, V2b) and (U2c, V2,) for the piece 29.  [0032] Typically, the atlas 6 is organized so that the textures which have different graphic properties are placed at distinct locations of the image 27.  For example, graphical properties denote values representative of the pixel intensities.  The list 22 comprises, for each vertex, the coordinates of this vertex expressed in a reference of the space.  In the example of Figure 2, we note A, B and C the vertices of the polygon 12.  The polygon 14 here has the vertices B, C and D 30 because, in this example, the polygons 12 and 14 are adjacent and the vertices B and C are common to the polygons 12 and 14.  The list 24 includes, for each polygon, an entry listing the vertices of the list 22 contained in this polygon.  For example, the entry of the polygon 12 in the list 24, denoted "P1", has the vertices "ABC".  The list 26 includes, for each polygon of the list 24, a set of UV coordinates which defines the piece of texture associated with this polygon.  For example, the entry "P1" associated with the polygon 12 comprises the coordinates, noted (u1, v1), of the three points which delimit the piece of texture 28 in the atlas 6.  The same goes for the entry, denoted P2, corresponding to the polygon 14.  The entry P2 is associated by list 26 with the coordinates (u2, v2) of the three points which delimit the piece of texture 29.  For example, the data 10 are compatible with the standard "OpenGL" (for "open graphics library" in English) well known to those skilled in the art.  FIG. 3 represents an exemplary device for transmitting the object 2 from a transmitter 40 to a receiver 42.  The transmitter 40 and the receiver 42 are able to exchange data with each other by means of a link 44 for exchanging data.  Generally, the transmitter 40 and the receiver 42 are separated from each other by several meters or several kilometers.  The link 44 is typically a link established through a long distance information network such as the World Wide Web.  The transmitter 40 includes a computing unit 46 which includes: - an information recording medium 48; a programmable electronic calculator 50; An interface 52 for exchanging data.  The support 48 includes the instructions necessary to perform the compression process of FIG. 4.  The calculator 50 reads and executes the instructions recorded on the support 48.  The interface 52 allows the exchange and the transfer of the data 10 on the link 44.  The transmitter 40 is for example a distribution server for a multimedia content.  The receiver 42 is here a client terminal such as a computer or a mobile communication device.  The receiver 42 is particularly suitable for displaying the object 2 ("rendering" in English).  It comprises here for that a graphic processor 48 and a computing unit 47, for example identical to the unit 46.  The transmitter 40 is able to compress the data 10 and then transmit them to the receiver 42 so that the latter decompresses them and displays the object 2 on a screen 49.  An example of data compression 10 will first be described, with reference to the flowchart of FIG. 4 and using FIGS. 1 to 7C.  This method is a method of progressive compression of the mesh ("progressive compression" in English).  First, in a step 60, the data 10 is acquired by the unit 46.  Then, during a step 62, the mesh 4 is shaped for the subsequent steps of the compression process.  These steps, and in particular simplification, will be described in more detail in the following.  This step comprises first a search operation 64, among the polygons of the mesh 4, first and second adjacent polygons that have different textures.  Two polygons are said to be adjacent if they have between them a first 40 and a second vertex in common and are interconnected by a common edge.  Here, two textures are said to be different if the intersection of the corresponding textured pieces in atlas 6 is a null set, i.e. the texture pieces are completely disjoint in the atlas 6.  For example, the texture pieces 28 and 29 are disjoint because the smallest distance between them is non-zero.  Here, the polygons 12 and 14 are adjacent and have in common the vertices B and C.  In addition, their respective textures 28 and 29 are different.  Also, in what follows, we will use the references 12 and 14 to designate generically such polygons.  Likewise, the references B and C will be used to designate the vertices in common with the polygons 12 and 14.  For example, the polygons 12 and 14 are identified by searching automatically, using the lists 24 and 26, polygons which are both adjacent to each other and which are associated by the list 26 to pieces of texture. disjoint.  If polygons 12, 14 have been identified, then, step 62 includes, for each pair of identified polygons, the separation of these polygons as follows.  In an operation 66, the vertices B and C are replaced in the polygon 14 by vertices B 'and C', said double vertices.  The vertices B 'and C' respectively occupy the same positions in space as the vertices B and C.  They have the same coordinates as these.  For example, the vertices B 'and C' are created in the list 22 and the BCD definition of the polynomial 14 in the list 24 is replaced by the definition B'C'D.  Thus, in the data 10 a new edge B'C 'is created and the edges BD and CD of the polygon 14 are replaced by edges, respectively, B'D and C'D.  Subsequently, the polygon 14 thus modified bears the reference 14 '(FIG. 5).  In this description, an edge that directly connects two vertices X and Y in a polygon has the reference "XY".  In an operation 68, at least two edges of zero length are created between, on the one hand, the vertices B and B 'and, on the other hand, the vertices C and C', so as to creating at least one intermediate polygon interposed between the polygons 12 and 14 '.  These edges BB 'and CC' connect between them the polygons 12, 14 '.  This connection is necessary to avoid the appearance of additional artefact at the junction of the polygons 12, 14 'as, for example, the appearance of a hole.  The face of this intermediate polygon has a null area so that it is not visible on a screen and creates no visible artifacts.  Here, since the polygons of the mesh 4 are necessarily triangles, they are two intermediate triangular polygons 67, 69 (FIG. 5), interposed between the polygons 12 and 14 ', which are created.  For example, an additional edge B'C is created to form these polygons 67 and 40 69.  Here, the polygon 67 is delimited by the edges BC, CB 'and BB' and the polygon 69 is delimited by the edges B'C ', CC' and CB '.  The new edges BB ', CC' and CB 'are created by adding in the list 24 the definition of these new polygons 67, 69.  For example, the polygons 67 and 69 have a texture of zero value and are therefore not associated with any piece of texture by the list 26.  [0054] To facilitate the reading of FIG. 5, the triangles 67 and 69 are drawn with a non-zero surface face whereas, as explained above, they have a zero surface.  Likewise, in this figure, the edges BB ', CC' are represented with a non-zero length whereas this is not the case.  The fact that a polygon has a face of zero area does not mean that it does not exist in the mesh 4, since any polygon is defined according to its vertices, especially in the list 24.  The same goes for the edges.  These operations 66 and 68 result in a corresponding modification of the lists 22, 24 and 26 of the data 10 to reflect the modifications of the mesh 4.  The duplication of the vertices B and C, by replacing them in the polygon 14 by the vertices B 'and C', makes it possible to separate the polygons 12 and 14 from one another.  This prevents the information that there is a difference in texture between these two polygons to be lost as soon as one of the vertices B and C is removed.  In addition, the fact that the vertices B 'and C' are connected to the vertices B and C by ridges prevents the polygons 12 and 14 'from moving relative to each other during the rest of the process.  Such a displacement would generate holes in the mesh 4, which would degrade the graphic quality of the object 2.  The zero value of the lengths of the edges and the area of the face of the polygons 67 and 69 cause this duplication of the vertices B and C not to result in a modification of the external graphic aspect of the mesh 4 and therefore of the object 2.  These edges make it possible to locally modify the connectivity of the mesh 4, so that subsequent simplification operations, which are based on connectivity information, do not result in these polyons disappearing too quickly.  When no more pairs of adjacent polygons of different texture are found in the modified data, then step 62 ends.  At the end of step 62, the mesh 4 does not have adjacent polygons that have different textures.  Then, during a step 70, the mesh 4 is simplified.  This simplification involves the removal of vertices, and therefore of polygons, from the mesh 4.  This simplification is intended to create new polygons that have a larger face than the deleted polygons.  These new polygons replace the deleted polygons.  This gives a simplified image that takes up less space in memory.  Step 70 comprises for this purpose: an identification operation 72, according to a predetermined criterion, vertices 40 to be deleted; An operation 74 for deleting the identified vertices and edges that connect these vertices to each other and to the other vertices of the mesh 4; a creation operation 76: new edges for connecting the vertices that have not been removed, thus creating the new polygons and new textures for covering the faces of the new polygons from the respective textures of the polygons deleted.  The identification during the operation 72 is for example carried out by selecting the vertices of the list 22 on the basis of connectivity criteria, such as the number of their nearest neighbors.  The operation 74 has the consequence of deleting a part of the polygons of the mesh 4.  In this example, step 70 is performed using the algorithm described in the following documents: 15 - "Rate-distortion optimization for progressive compression of 3D mesh with color attributes", Ho Lee et al, The Visual Computer, vol.  28, p.  137-153, Springer-Verlag, May 2011, DOI: 10. 1007 / s00371-011-0602-y.  - P.  Alliez et al.  "Progressive compression for lossless transmission of triangle meshes", ACM Proceedings of SIGGRAPH, p.  195-202, 2001.  [0064] FIGS. 6A to 6C illustrate the operation of this algorithm on a portion 80 of the mesh 4.  This algorithm works in two stages.  First, operations 72 to 74 are performed during a so-called "conquest" phase.  The vertices of the mesh 4 are automatically traversed step by step, moving along the edges 25 of the mesh 4, in the manner of a graph.  The order in which vertices are traversed is defined by a predefined graph traversed in English.  For each vertex, we determine what is the valence of this vertex, that is to say the number of immediate neighbors S 'of this vertex to which it is directly connected by edges.  If this vertex S has a valence lower than a predetermined threshold, then it is deleted.  The edges that linked this vertex S to its immediate neighbors are then removed.  These edges bear the references SS 'in FIG. 6A.  The new edges 80, 81, 82 (FIG. 6B) are created to replace the edges SS '.  In a second, so-called cleaning phase, the algorithm advantageously eliminates certain surplus vertices 83 as well as the edges connecting these vertices and creates new edges in replacement, so as to obtain new polygons 84 having a regular shape.  The operation 76 is here partly carried out during this phase, since edges are created there.  These operations 72, 74 and 76 are here reflected by corresponding modifications of the lists 22, 24 and 26 and of the atlas 6 to reflect the modifications made to the mesh 4.  Advantageously, during an operation 89, are recorded 5 so-called incremental data or refinement.  These data indicate which vertices and edges were removed as well as those created in step 70.  For example, these data include a list of vertices removed in step 70, as well as a list giving, for each of these deleted vertices, the set of neighboring vertices to which this deleted vertex was directly connected by an edge. .  The incremental data makes it possible, when used in a decompression process, to perform the operations opposite to those of step 70.  This makes it possible to reconstruct the mesh as it was before application of step 70 from the simplified mesh obtained at the end of this step 70 and this incremental data.  Advantageously, these data also comprise information enabling the piece of texture associated with each reconstructed polygon to be retrieved without loss of information.  FIG. 7A represents a simplified version of object 2, also called simplified object 94, which is available at the end of step 70 after the simplification of object 2.  Object 94 has fewer vertices and polygons than object 2.  Because of this lower resolution, the data representative of this object 94 is smaller in size than the data 10.  The transmission of the object 94 is thus facilitated.  In this example, the simplification is repeated several times to obtain a higher compression ratio.  In this case, the method further comprises, at the end of step 70, an additional simplification step 91.  Step 91 is for example identical to step 70, except that it is performed on object 94 so as to simplify this object 94.  Typically, during each new iteration of step 91, the selection criteria for the vertices to be deleted are widened to remove new vertices.  Thus, at the end of step 91, a final version of object 2, referred to as final object 96, is obtained as shown in FIG. 7B.  Object 96 has fewer vertices and polygons than object 94.  It corresponds to a version of the object 2 more simplified than is the object 94.  The incremental data generated during this execution of step 91 is also recorded.  Finally, during a step 92, the object 96 is transmitted on the interface 50.  The present method is particularly advantageous for reducing the appearance of graphic artifacts on objects 94, 96 upon compression.  Indeed, the progressive compression methods conventionally used do not take into account the differences in texture between adjacent polygons.  In such a conventional method, the vertices B and C of the polygons 12 and 14 can be quickly removed during operation 74.  The polygons 12 and 14 are then deleted and replaced by 3028990 14 new polygons.  Then, the texture of the polygons is replaced by a new texture determined from the points of the texture pieces 28 and 29.  For example, it frequently happens that one of these new polygons is associated with a new piece of texture defined by two points of the pieces 29 and a point of the piece 28.  This new piece includes a portion of the image 27 located between the pieces 28 and 29.  But this portion is often completely different from pieces 28 and 29.  This makes appear at the location of the polygons 12 and 14, in the simplified object, a piece of very different texture and therefore particularly visible.  This gives rise to graphic artifacts that are particularly visible because of the difference in texture between the polygons 12 and 14.  Figures 8A and 8B illustrate, respectively, objects 94 'and 96'.  The object 94 'is a simplified object obtained at the end of a compression process applied to the object 2 and identical to the method of FIG. 4 (without the step 91), but in which the step 62 has not been executed.  Likewise, the object 96 'is a final object obtained after execution of the compression process on the object 2 during which the steps 70 and 91 have been applied, but in which step 62 has not been executed.  The object 94 'is identical to the object 94, except that it comprises a graphic artifact 100.  Likewise, the object 96 'is identical to the object 96 except that it has a graphic artifact 102.  These artifacts 100 and 102 correspond to polygons whose texture information has been partially or totally lost in step 70 or 91.  This loss of information results from the removal of adjacent polygons that had different textures.  Since these polygons were not separated in step 62 prior to the simplification step, they were removed in operation 74.  Conversely, when the method of FIG. 4 is applied, the information according to which the adjacent polygons 12 and 14 have different textures is not lost until all the vertices B, B ', C and They have not been deleted.  This loss of information can therefore occur only after a much larger number of iterations of steps 70 and 91.  Thus, either the visual artifacts do not appear, or they appear at a much more advanced stage of compression.  However, as will be understood from the following, the more the visual artefact appears at an advanced stage of compression, the faster it disappears during decompression.  With the method of FIG. 4, during decompression, the artifacts are thus removed or much more ephemeral.  FIG. 9 represents an information support 104 containing a set 106 of data representative of the object 96 '.  posol The simplified object is subsequently transmitted by the transmitter 40 to the receiver 42, to be displayed.  Likewise, incremental data respectively associated with each execution of step 70 or 91 are also transmitted from transmitter 40 to transmitter 42.  This transmission is for example carried out sequentially.  FIG. 10 illustrates the decompression method for reconstructing the object 2.  This process is for example that described in the patent application FR2998685.  In a step 110, the data representative of the object 96 are acquired first by the receiver 42, following their transmission from the transmitter 40.  The receiver 42 then immediately displays the received object 96.  Then, during a step 112, the receiver 42 automatically reconstructs the object 94 from the object 96 and using the incremental data that was generated during the step 91, by performing reverse operations from those performed in this step 91.  These incremental data are typically acquired by the receiver 42 after receiving the representative data of the object 96 and, often, after the display of the object 96 on the screen 49.  Then, from these data, the receiver 42 modifies the object 96 to add the vertices of the mesh which were removed during the step 91.  This leads to the removal of some polygons from the mesh of object 96 to replace them with replacement polygons, which are more numerous and have a smaller area.  Thus, in the object 96, the receiver 42: 20 - adds the vertices removed during the simplification of the step 91, - replaces the edges of the mesh which had been created during the step 91, by additional edges that connect the vertices added to the existing vertices of object 96, so as to create additional polygons; - Creates additional textural pieces, for these additional polygons, from the respective textures of the replaced polygons and incremental data.  At the end of step 112, the receiver 42 has reconstructed the object 94.  The object 94 is then displayed in place of the object 96 on the screen 49.  Then, during a step 114, the receiver 42 reconstructs the object 2 from the reconstructed object 94 and using the incremental data that was generated during the step 70.  This step 114 is for example identical to step 112, except that it applies to object 94 rather than object 96.  Thus, object 2 is progressively reconstructed, by successive refinements from incremental data, with increasing accuracy, as and when received, until a level of detail identical or similar to that he presented before compression.  Intermediate simplified objects, such as object 94, are displayed as soon as they are rebuilt.  The receiver 42 thus displays, as it receives the incremental data and implements the reconstruction steps, objects whose accuracy is increasing.  This method 40 is useful when the flow rate of link 44 is limited or subject to significant variations.  This prevents the display of the object on the receiver 42 is disturbed by a phenomenon of lag ("lag" in English).  This also limits the waiting time for the object to be displayed (even in a simplified version) by the receiver 42, compared to the case where it would be necessary to wait for all the compressed object 2 to be transmitted to the receiver 42 before you can start decompression and then display it.  Since, during the decompression, the simplified objects are displayed in the reverse order of their creation by the compression, the more the visual artefact appears late during the compression, the faster it disappears during the decompression and 10 therefore the progressive display of the uncompressed object.  Advantageously, during a step 116, the mesh of the reconstructed object 2 is shaped to remove the vertices and edges created during step 62.  This step comprises here: the replacement of the vertices B 'and C' by, respectively, the vertices B and C and the deletion of the edges BB ', CC' and B'C.  Thus, the polygons 67, 69 are deleted and the polygons 12, 14 find the configuration they had initially in the object 2.  Many other embodiments are possible.  The object 4 may be different.  It can be any object that can be represented as a mesh.  Similarly, atlas 6 can be chosen differently.  The polygons may have a shape other than a triangle.  For example, polygons have a parallelepiped shape.  The data 10 may be recorded in a format compatible with another graphic standard, such as Direct3D.  The transmitter 40 may be different.  For example, the unit 46 may be distinct from the transmitter 40.  The compression of the object 2 is therefore not performed by the transmitter 40, the latter then only for the purpose of transferring the data to the receiver 42.  The receiver 42 may be different.  It can for example be a tablet, a mobile phone, a TV.  The display of the object 2 can also be achieved independently of the reception.  The term "receiver" 42 thus encompasses two distinct devices, one receiving object 2 and the other displaying object 2.  Step 62 may be performed otherwise.  In particular, the operation 64 may be performed differently to identify adjacent polygons.  For example, FIG. 11 illustrates another method for identifying the polygons 12, 14 in a portion of the mesh 4.  According to this method, the edges of the mesh are automatically scanned by selecting the vertices of the mesh one after the other in a predetermined order.  For each selected vertex 150, all the immediately adjacent vertices 152, 153 of this vertex 150 are scanned, turning around the vertex 150 in a predefined direction, meaning here identified by the arrow 154.  These neighboring vertices are defined as the vertices directly connected to the vertex 150 by an edge.  For simplicity, only vertices 152 and 153 are numerically referenced.  Then, if they exist, vertices 152 and 153 are determined having another immediately adjacent common vertex 156 other than the vertex 150.  If so, it indicates that the vertices 152, 153 are common to several polygons.  In this example, adjacent polygons 158 and 160 are identified.  Polygon 158 has vertices 150, 152, and 153.  Polygon 160 has vertices 152, 153, and 156.  Next, it is checked whether the textured pieces associated with the polygons 158 and 160 are different.  [cm 00] Alternatively, it is the vertices B and C of the polygon 12 which are replaced by the vertices B 'and C'.  [00101] Step 91 can be omitted.  In this case, the object 94 is the final object and it is it that is transmitted.  Step 112 is then omitted from the decompression process.  According to other variants, the step 91 is repeated several times, so as to increase the compression ratio of the object 2 before its transmission.  The decompression method then comprises a number of identical steps in step 112 equal to the number of times in which step 91 has been repeated.  [00103] In another variant, when step 91 is executed at least once, then step 62 is repeated, for example before each execution of step 91.  If the polygons 67 and 69 have been deleted during an execution of the step 91, they can be recreated before performing step 91 again.  This limits the risk of polygons 12 and 14 being removed in this step 91.  In this case, step 116 may be repeated several times.  For example, step 116 is applied during decompression after each step 112 or 114. .  Step 70 may be performed differently.  For example, the algorithm described in the following document is used: Maglo, A. , Courbet, C. , Alliez, P.  .  "Progressive compression of manifold polygon meshes", Computers and Graphics, 2012, DOI: 10. 1016 / j. cag. 2012. 03. 023.  [00105] Step 116 may be omitted.  This step can also be applied after step 112 to clean the mesh of the reconstructed object 94 before step 114 is applied.  [00106] In another embodiment, during step 62, the vertices B 'and C', as well as the vertices B and C, are marked during an operation 180 (Figure 4), indicating that they should preferably not be deleted in step 70.  This marking consists for example of a predefined value of an additional data bit added in the list 22 for each vertex.  Thus, during step 70, it is verified during operation 72 if the vertex has such a marking.  For example, we make a first run of all 40 vertices of the mesh.  If an unmarked vertex is met and it satisfies the 3028990 18 criteria to be deleted, then it is deleted.  On the other hand, if it is marked, then one does not delete it from the beginning and one indicates in a specific list which will be consulted only during a second course of the vertices of the mesh.  If step 70 can end without it being necessary to delete it, then that vertex is not deleted.  If it is impossible not to delete this vertex (for example because its deletion is imposed to preserve some properties of regularity of the mesh) then this vertex is suppressed.  In other words, it is delayed until it can not be done otherwise.  The marking may be reset after each application of step 70 and 91, to take into account the mesh modifications induced by the simplification.  In an extreme case, the removal of a vertex so marked is forbidden, which ensures that no visual artefact will appear.  [00107] Finally, other methods are possible to identify pieces of different textures.  For example, in a simplified version, pieces of texture are different as soon as they have no point in common in the image 27.  In another variant, pieces of textures are different only if the minimum distance d between these two pieces is strictly greater than a predetermined threshold.  This predetermined threshold may be zero or strictly greater than zero.  20 [cm 08] It is also possible to compare features that are representative of the texture to decide whether or not different textures are different.  For example, for each piece of texture, this characteristic is calculated and then if the difference between the value of this characteristic for a first piece and a second piece is greater than a predetermined threshold, then these pieces of texture are said to be different.  The characteristic can be the median value or the maximum or the minimum of a histogram of the colors contained in the piece.  The characteristic may be a magnitude representative of the graphic pattern on the texture, such as a fractal dimension or a Hurst exponent.  This latter procedure based on representative features does not use the positions of the points delineating the textured pieces in the image 27.
    权利要求:
    Claims (11)
    [0001]
    REVENDICATIONS1. A method of compressing data representative of a digital three-dimensional object, comprising: - the acquisition (60) of data (10) representative of a three-dimensional object (2), these data comprising: a mesh (4) formed of a plurality of planar polygons contiguous with each other, each polygon comprising: a plurality of vertices (A, B, C) interconnected by edges which delimit a face of the polygon, and, - a texture (28) covering the face of the polygon; an atlas of texture (6) in which are included all the textures of the polygons of the mesh; the simplification (70) of the mesh, by removing vertices from the mesh, which eliminates polygons and instead creates new polygons having a larger face, comprising: the identification (72), according to a predetermined criterion, vertices to be suppressed, - the deletion (74) of the identified vertices and the edges connecting these vertices to other vertices of the mesh, so as to eliminate these polygons comprising these vertices and these edges, - the creation (76) : - new edges to connect vertices that have not been removed; - new textures, for these new polygons, from the respective textures of the deleted polygons; characterized in that, before deleting vertices, the method comprises: - identifying (64), in the mesh, first (12) and second (14) adjacent polygons that have different textures, two polygons being said to be adjacent 30 if they present a first (B) and a second (C) vertex in common interconnected by a common edge, then: the replacement (66) in the second polygon of the first and second vertex in common by, respectively, third (B ') and fourth (C') vertices, these third and fourth vertices occupying the same positions in space as, respectively, the first (B) and second (C) vertices; the creation (68) of two edges (BB, CC ', B'C) of zero length between, respectively, the first and third vertices and the second and fourth vertexes and therefore at least one intermediate polygon (70, 72). zero surface interposed between the first and second polygons.
    [0002]
    2. Method according to claim 1, wherein: the method comprises, before deleting vertices, the marking (180) of the first, second, third and fourth vertices; when simplifying (70) the mesh, the first, second, third and fourth vertices which have been the subject of said marking are not identified as vertices to be deleted even if they satisfy the predetermined criterion. 10
    [0003]
    3. Set (106) of data representative of a compressed three-dimensional digital object, these data comprising: a simplified mesh, formed of a plurality of polygons and obtained by means of a compression method according to claim 1 or 2; ; Incremental decompression data, which data includes a list of vertices and edges that have been removed from the mesh when performing said compression method; characterized in that at least one of the simplified mesh and incremental decompression data encodes first, second, third and fourth vertexes of the mesh such that the first and third vertices are interconnected by an edge of zero length and the second and fourth vertices are interconnected by an edge of zero length.
    [0004]
    4. Information recording medium (104), characterized in that it comprises the set of data representative of a compressed digital three-dimensional object according to claim 3.
    [0005]
    A method of decompressing data representative of a compressed three-dimensional digital object, comprising: acquiring (110) a set of data representative of a compressed digital three-dimensional object according to claim 3; - the reconstruction (112) of a more complex mesh from the simplified mesh and incremental data, this reconstruction, comprising the creation of: - additional vertices in the simplified mesh from the acquired incremental data and the replacement of edges of the mesh simplified by additional edges connecting these new vertices to existing vertices, so as to remove polygons of the simplified mesh and replace them with additional polygons having a smaller area; additional textures, for these additional polygons, from the respective textures of the replaced polygons and incremental data, characterized in that the more complex reconstructed mesh comprises the first and third vertices interconnected by an edge of zero length and the second and fourth vertices interconnected by an edge of zero length.
    [0006]
    The decompression method according to claim 5, wherein the reconstructed mesh is shaped (116) by: replacing the third and fourth vertices of the second polygon by, respectively, the first and second vertices and; edges of zero length between the first and third vertices on the one hand and the second and fourth vertices on the other hand. 15
    [0007]
    A method of transmitting data representative of a digital three-dimensional object between a transmitter (40) and a receiver (42), the method comprising: acquiring data representative of a digital object by the transmitter; compressing the acquired data by means of a compression method according to claim 1 or 2; the transfer of the compressed data from the transmitter to the receiver by means of a link (44) for data exchange; the decompression, by the receiver, of the compressed data transferred, by means of a decompression method according to claim 5 or 6.
    [0008]
    The method according to any one of claims 1 to 2 and 5 to 7, wherein two textures (28, 29) are said to be different if their intersection in the texture atlas (6) is equal to the null set. . 30
    [0009]
    9. A method according to any one of claims 1 to 2 and 5 to 7, wherein two textures (28, 29) are said to be different if the shortest distance d between these two textures in the texture atlas ( 6) is strictly greater than a predetermined threshold. 35
    [0010]
    Information recording medium (48), characterized in that it comprises instructions for the execution of a method according to any one of claims 1 to 2 and 5 to 9 when these instructions are executed. by an electronic calculator. 3028990 22
    [0011]
    11. Electronic calculator (46) for the implementation of any one of claims 1 to 2 and 5 to 9, the electronic calculator being programmed to: - acquire (60) data (10) representative of a three-dimensional object (2), these data comprising: a mesh (4) formed of a plurality of planar polygons contiguous with each other, each polygon comprising: a plurality of interconnected vertices (A, B, C) edges defining a face of the polygon, and a texture (28) covering the face of the polygon; an atlas of texture (6) in which are included all the textures of the polygons of the mesh; simplifying (70) the mesh, by removing vertices from the mesh, which deletes polygons and instead creates new polygons having a larger face, by: - identifying (72), according to a predetermined criterion vertices to delete, - removing (74) identified vertices and edges connecting these vertices to other vertices of the mesh, so as to remove these polygons with these vertices and edges, 20 - creating (76): - new edges to connect vertices that have not been removed; - new textures, for these new polygons, from the respective textures of the deleted polygons; characterized in that the computer is programmed to, before deleting vertices,: - identifying (64), in the mesh, first (12) and second (14) adjacent polygons that have different textures, two polygons being said adjacent if they have a first (B) and a second (C) vertex in common interconnected by a common edge, then: - replacing (66) in the second polygon of the first and second vertex in common by, respectively, third (B ') and fourth (C') vertices, these third and fourth vertices occupying the same positions in space as, respectively, the first (B) and second (C) vertices; creating (68) two edges (BB, CC ', B'C) of zero length between, respectively, the first and third vertices and the second and fourth vertices and thus creating at least one intermediate polygon (70, 72) of zero surface interposed between the first and second polygons.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3221846B1|2022-02-16|Methods for compressing and decompressing data representing a digital three-dimensional object and information-recording medium for recording information containing said data
    EP1292921B1|2005-11-16|Refinement of a three-dimensional triangular mesh
    EP2304686B1|2018-04-18|Method and device for filling in the zones of occultation of a map of depth or of disparities estimated on the basis of at least two images
    EP3216219A1|2017-09-13|Method for producing animated images
    EP1413140B1|2006-09-27|Method for estimating the motion between two images with management of mesh overturning, and corresponding coding method
    WO2010003844A1|2010-01-14|Method for the real-time composition of a video
    FR2933520A1|2010-01-08|METHOD AND DEVICE FOR RESTORING A VIDEO SEQUENCE
    FR3083950A1|2020-01-17|METHOD FOR VIEWING GRAPHIC ELEMENTS FROM AN ENCODE COMPOSITE VIDEO STREAM
    EP1141899B1|2005-04-20|Method for simplifying a source mesh, taking into account the local curvature and the local geometry, and corresponding uses
    FR3105687A1|2021-06-25|PROCESS FOR ENCODING A DIGITAL IMAGE FOR COMPRESSION
    EP1864260B1|2011-01-05|Method for image reconstruction in a vector graphic
    EP2204775B1|2011-10-05|Incremental polygon triangulation for digital display
    FR2929431A1|2009-10-02|METHOD AND DEVICE FOR CLASSIFYING SAMPLES REPRESENTATIVE OF AN IMAGE DIGITAL SIGNAL
    FR2994500A1|2014-02-14|METHOD OF OPTIMIZING THE EXECUTION OF A GRAPHIC PIPELINE
    EP2737452B1|2019-04-10|Method for encoding an image after resizing by deleting pixels and image transmission method between a transmitter and receiver
    EP1069531A1|2001-01-17|Integration of constraint lines in a Delaunay triangulation
    EP1121665B1|2003-03-26|Source mesh coding method optimising vertex position resulting from edge fusion, and corresponding applications
    BE1021546B1|2015-12-11|METHOD AND SYSTEM FOR STORING WAVEFORM DATA.
    FR3096538A1|2020-11-27|Multi-view video data processing method and device
    FR3058547A1|2018-05-11|METHOD AND DEVICE FOR RESTORING DIGITAL IMAGE
    FR3083662A1|2020-01-10|ROBUST COMPRESSION AND DECOMPRESSION OF DIGITAL IMAGES
    EP2192555B1|2012-01-18|Anzeige von parametrisierten Daten
    WO2015055902A1|2015-04-23|Method of encoding and decoding a sequence of video images with predetermined modes of filling for the pixels of occluded parts in the images
    FR2837949A1|2003-10-03|Method for the visualization of a semi-real geographic image, uses local table created in memory to hold interpolated and extrapolated values derived from Geographic Information System database
    同族专利:
    公开号 | 公开日
    WO2016079430A1|2016-05-26|
    FR3028990B1|2018-01-19|
    JP6689269B2|2020-04-28|
    US10192326B2|2019-01-29|
    EP3221846B1|2022-02-16|
    EP3221846A1|2017-09-27|
    JP2017539011A|2017-12-28|
    US20170365069A1|2017-12-21|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20040164982A1|2003-02-24|2004-08-26|Minolta Co., Ltd.|Method and apparatus for editing three-dimensional model, and computer readable medium|
    EP0889440B9|1997-06-30|2004-03-17|Sun Microsystems, Inc.|Method and apparatus for geometric compression of three-dimensional graphics|
    WO1997006512A2|1995-08-04|1997-02-20|Microsoft Corporation|Method and system for rendering graphical objects to image chunks and combining image layers into a display image|
    US6208347B1|1997-06-23|2001-03-27|Real-Time Geometry Corporation|System and method for computer modeling of 3D objects and 2D images by mesh constructions that incorporate non-spatial data such as color or texture|
    US6167159A|1998-04-30|2000-12-26|Virtue Ltd.|Triangle mesh compression|
    FR2781908B1|1998-08-03|2000-09-29|France Telecom|METHOD OF CODING A MESH WITH A REDUCED NUMBER OF FACETS, PROCESS FOR RECONSTRUCTION AND CORRESPONDING APPLICATIONS|
    FR2794881B1|1999-06-10|2001-08-17|France Telecom|CODING METHOD OF A CONQUEST MESH ON THE ARETES, PREFERRED A FULL SUMMIT AS A PIVOT, AND CORRESPONDING DECODING METHOD|
    JP3530125B2|2000-09-27|2004-05-24|彰 川中|Method and apparatus for forming structured polygon mesh data, and storage medium|
    US20110046923A1|2008-04-18|2011-02-24|Electronics And Telecommunications Research Institute|Apparatus and method for low-complexity three-dimensional mesh compression|
    GB2477368B|2010-10-11|2011-12-14|Visual Technology Services Ltd|Triangle mesh compression|
    US8736603B2|2011-11-02|2014-05-27|Visual Technology Services Limited|Compression of texture rendered wire mesh models|
    FR2998685B1|2012-11-27|2016-01-08|Inst Nat Sciences Appliquees Lyon|METHOD FOR TRANSMITTING BETWEEN TWO DATA TERMINALS REPRESENTATIVE OF A THREE-DIMENSIONAL OBJECT|
    EP2833326B1|2013-07-30|2016-04-06|Dassault Systèmes|Lossless compression of a 3D mesh including transforming of the mesh to a image|US10389994B2|2016-11-28|2019-08-20|Sony Corporation|Decoder-centric UV codec for free-viewpoint video streaming|
    KR101919085B1|2017-04-19|2018-11-16|에스케이텔레콤 주식회사|Apparatus and method for simplification of 3D mesh data|
    US10459706B1|2018-03-12|2019-10-29|State Farm Mutual Automobile Insurance Company|System and methods for simplifying three-dimensional models|
    US10839590B2|2018-07-16|2020-11-17|Microsoft Technology Licensing, Llc|Gutter space padding for texture atlases|
    GB2583513A|2019-05-02|2020-11-04|Sony Interactive Entertainment Inc|Apparatus, system and method for data generation|
    法律状态:
    2015-11-23| PLFP| Fee payment|Year of fee payment: 2 |
    2016-05-27| PLSC| Publication of the preliminary search report|Effective date: 20160527 |
    2016-11-27| PLFP| Fee payment|Year of fee payment: 3 |
    2017-11-21| PLFP| Fee payment|Year of fee payment: 4 |
    2019-11-20| PLFP| Fee payment|Year of fee payment: 6 |
    2020-11-20| PLFP| Fee payment|Year of fee payment: 7 |
    2021-11-30| PLFP| Fee payment|Year of fee payment: 8 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1461318A|FR3028990B1|2014-11-21|2014-11-21|METHODS OF COMPRESSING AND DECOMPRESSING REPRESENTATIVE DATA OF A DIGITAL THREE DIMENSIONAL OBJECT AND INFORMATION RECORDING MEDIUM CONTAINING THE SAME|
    FR1461318|2014-11-21|FR1461318A| FR3028990B1|2014-11-21|2014-11-21|METHODS OF COMPRESSING AND DECOMPRESSING REPRESENTATIVE DATA OF A DIGITAL THREE DIMENSIONAL OBJECT AND INFORMATION RECORDING MEDIUM CONTAINING THE SAME|
    PCT/FR2015/053122| WO2016079430A1|2014-11-21|2015-11-18|Methods for compressing and decompressing data representing a digital three-dimensional object and information-recording medium for recording information containing said data|
    EP15805595.4A| EP3221846B1|2014-11-21|2015-11-18|Methods for compressing and decompressing data representing a digital three-dimensional object and information-recording medium for recording information containing said data|
    US15/528,275| US10192326B2|2014-11-21|2015-11-18|Methods for compressing and decompressing data representing a digital three-dimensional object and information-recording medium for recording information containing said data|
    JP2017527657A| JP6689269B2|2014-11-21|2015-11-18|Method for compressing and decompressing data representing a digital three-dimensional object, and information recording medium for recording information including the data|
    [返回顶部]