Abstract: The present principles relates to a method and device for encoding an input colored point cloud representing the geometry and colors of a 3D object. The method comprises: —selecting (100) at least one face (Fi,j) of at least one cube (Cj) of an octree-based structure of projection according to at least one orthogonal projection of the point cloud onto said at least one face; and —encoding (120, 130) a pair of one texture image (TIi,j) and one depth (DIi,j) image per selected face (Fi,j) of a cube (Cj) by orthogonally projecting the part of the point cloud included in said cube (Cj) onto said selected face (Fi,j).

Abstract: The present principles relates to a two-steps approach for encoding and decoding the geometry of a point cloud. In a first step, an octree-based structure is obtained by splitting recursively a cube encompassing the point cloud until the leaf cubes associated with the leaf nodes of said octree-based structure reach down an expected size.

Abstract: Methods and devices for attribute coding for point clouds. In a top-down coding process a predicted geometry-weighted sum of attributes is determined for each occupied sub-volume of a volume based on a prediction operation that uses at least a geometry-weighted sum of attributes for the volume. The prediction operation involves up-sampling data from the preceding (parent) depth of the coding tree. The predicted geometry-weighted sums of attributes and corresponding original geometry-weighted sums of attributes are subtracted, in the attribute domain or in a transform domain to result in residual coefficients. The AC residual coefficients are coded in the bitstream. The transform used has a DC coefficient property, whereby the DC coefficient from transform of the geometry-weighted sums of attributes for the set of child sub-volumes is the geometry-weighted sum of attributes for the volume.

Abstract: A decoding method is disclosed that comprises:—decoding (S120) a stream to obtain a decoded standard dynamic range picture and colour metadata associated with the standard dynamic range picture, wherein the colour metadata are representative at least of characteristics of a high dynamic range picture associated with the standard dynamic range picture; and—reconstructing (S130) a high dynamic range access the input stream(s) picture from the decoded standard dynamic range picture and from the colour metadata.

Abstract: The present principles a method and a device for encoding the colors of a point cloud (IPC) representing a 3D object. The method is characterized in that the method comprises: —obtaining (210) a set of points (SP) approximating the geometry of said point cloud (IPC); —encoding (220) a color information data indicating that the color of at least one of the point of said set of points is encoded in a bitstream and that the colors of the other points of said set of points are not encoded in the bitstream. The present principles also relates to a method and a device for decoding the colors of a point cloud.

Abstract: A method of color mapping a video signal represented in a first color volume from color mapping data to be applied on a video signal represented in a second color volume is disclosed. The method comprises: color mapping (14) said video signal represented in a first color volume from said first color volume into said second color volume in the case where said first and second color volumes are different; and color mapping (16) said color mapped video signal based on said color mapping data.

Abstract: A method and device for modifying attributes of one or more points of a 3D scene, the attributes comprising depth attribute and texture attribute stored respectively in a depth image and a texture image. To reach that aim, at least a first point of the 3D scene is identified, the at least a first point having depth information inconsistent with regard to depth information associated with a neighborhood of the at least a first point in the depth image. It is then determined whether the attributes of the at least a first point are correctable by analyzing neighborhood of the at least a first point in the depth image and in the texture image. For each first point having correctable attributes, the attributes stored in the depth image and/or texture image are modified according to the neighborhood in the depth image and/or in the texture image.

Abstract: The present disclosure generally relates to a method and device for obtaining three color difference components from three components C1, C2 and C3 representing colors picture data in an input color space, a first of said color difference components being defined from the component C1, a second of said color difference components being defined from the first color difference component and the component C2, and a third of said color difference components being defined from the first color difference component and the component C3. The method is characterized in that it further comprises: —obtaining a first component by applying a process to the component C1; —obtaining said first color difference component by applying the inverse process to the first component.

Abstract: Significance flags in advanced video compression systems are coded using contexts adaptive to the last N significance flags coded taken in a scanning order. One embodiment uses the last N significance flags in scanning order as a predictor to determine which of a plurality of sets of significance flag contexts to use for coding subsequent significance flags. A second embodiment uses the last N significance flags in scanning order as a predictor in order to modulate the probability value associated with significance flag contexts that are used to code significance flags for future coding.

Abstract: A method and device for generating points of a 3D scene from a depth image. To reach that aim, depth information associated with a current pixel is compared with depth information associated with pixels spatially adjacent to the current pixel in the depth image. When the difference of depth between the current pixel and an adjacent pixel is greater than a first value and less than a second value, at least an additional point of said 3D scene is generated, in addition to a current point associated with the current pixel of the depth image, the number of additional points depending on the depth difference.

Abstract: This method for inter-predictive encoding of a time-varying 3D point cloud including a series of successive frames divided in 3D blocks into at least one bitstream comprises encoding (20) 3D motion information including a geometric transformation comprising rotation information representative of a rotation transformation and translation information representative of a translation transformation, wherein the translation information comprises a vector ?T representing an estimation error of the translation transformation.

Abstract: This method for up-sampling a point cloud representing a 3D object, comprises:—detecting (2) points belonging to at least one under-sampled region of the point cloud on the basis of at least one desirable sampling rate (K);—obtaining (12), for each detected point, an associated tangent plane;—inserting (14) in the point cloud at least one neighboring point of each detected point if a distance between the neighboring point and the tangent plane associated with the detected point is less than a first threshold.

Abstract: The present principles relate to a method and device for encoding the geometry of a point cloud representing a 3D object. The method comprises: a) obtaining (210) at least one candidate octree-based structure (Ok) with a maximum depth level (?k) from the points of said point cloud, each candidate octree-based structure (Ok) approximating the geometry of the point cloud by points associated with leaf cubes, each leaf cube being associated with a leaf node of said candidate octree-based structure (Ok); b) obtaining (230) a coding cost (Ck) for said at least one candidate octree-based structure (Ok) with a maximum depth level (?k); c) encoding (250) an optimal octree-based structure (Oopt) with a maximum depth level (?opt) obtained by optimizing (240) said coding cost obtained for said at least one candidate octree-based structure (Ok); and d) encoding (260) a depth information data (DIDopt) representative of the maximum depth level (?opt) of said optimal octree-based structure (Oopt).

Abstract: Method and device for encoding a point cloud that represents a three-dimensional (3D) object. One or more groups of temporally successive pictures are obtained. Each picture of the one or more groups comprises a first set of images, the images being spatially arranged in a same manner in each picture of the one or more groups. A second set of projections is associated with the one or more groups, a unique projection being associated with each image in such a way that a same projection is associated with only one single image and all projections are associated with the images. A first information representative of the projections is encoded. The point cloud may then be encoded according to the obtained pictures. A corresponding method and device for decoding a bitstream that comprises data representative of the encoded picture(s) are also described.

Abstract: A method for determining a luminance value used to process color samples in a decoder or encoder is provided comprising determining (33, 44) a luminance sample associated with a chroma sample position in a common space; and applying (34, 492) the determined luminance sample in a process that depends on luminance that is being applied to the chroma sample.

Abstract: This method for encoding a 3D point cloud comprises: —segmenting the point cloud into first patches (2); —determining inverse projection parameters of the patches (4); —projecting the first patches to planes using the inverse projection parameters (10); —encoding the planes (12); —determining sparse points in the point cloud not belonging to any patch (HC, 18); and—encoding the sparse points using an octree-decomposition (20).

Abstract: The present principles relate to an architecture of a coding scheme that encodes depth images (and optionally texture images) obtained by projecting a point cloud on surfaces associated with volumes of a structure of projection. A global octree-based structure is also obtained from the point cloud and an inverse-projected point cloud obtained by inverse projecting the decoded depth images (and optionally decoded texture images).

Abstract: Method and devices for coding point cloud data using a planar coding mode. The planar coding mode may be signaled using in a planar mode flag to signal that a volume is planar. A planar volume has all of its occupied child nodes on one side of a plane bisecting the volume. A planar position flag may signal which side of the volume is occupied. Entropy coding may be used to code the planar mode flag and/or the planar position flag. Context determination for coding may take into account one or more of whether a parent volume containing the volume is planar in occupancy, occupancy of a neighbouring volume at a parent depth, distance between the volume and a closest already-coded occupied volume at a same depth as the volume, plane position, if any, of the closest already-coded occupied volume, and a position of the volume within the parent volume.

Abstract: A method is disclosed that comprises mapping a high-dynamic range luminance picture to a standard-dynamic range luminance picture based on a backlight value Bac associated with the high-dynamic range luminance picture.

Abstract: The present principles relates to a method and a device for encoding the colors of a point cloud wherein the method comprises: —assigning a specific number (LNk) per leaf node of an octree-based structure representing the geometry of the point cloud, a specific number (LNk) being assigned to a leaf node by assigning, recursively, an orientation ({?0, . . . , ?7}) (201) and a local number ({n0, . . . , n7}) (202) to each of eight child nodes of said octree-based structure according to an orientation (?p) previously assigned to a parent node and according to a specific child node numbering order; —ordering (200), the points of the point cloud by scanning the leaf nodes of the octree-based structure according to said assigned specific numbers; —assigning (210) colors of ordered points of the colored point cloud to pixels of an image according to a specific 2D path; and—encoding (220) the image. The present principles also relates to a method and device for decoding a point cloud.