WO2021249867A1

WO2021249867A1 - A method and apparatus for encoding and decoding volumetric video as partitioned patch atlases

Info

Publication number: WO2021249867A1
Application number: PCT/EP2021/064924
Authority: WO
Inventors: Frédérique Humbert; Charles Salmon-Legagneur; Charline Taibi; Rémi HOUDAILLE
Original assignee: Interdigital Vc Holdings France, Sas
Priority date: 2020-06-12
Filing date: 2021-06-03
Publication date: 2021-12-16

Abstract

Methods, apparatus and streams are disclosed for encoding, transmitting and decoding volumetric video. Volumetric video is encoding in the form of a patch atlas sequence. The patch atlas is generated, and a sub-set of patches is selected to be partitioned, for instance on a size criterion. Metadata are associated with the atlas in a data stream to indicate which and according to which partitioning mode patches are encoded. At the rendering side, a level of sampling level is selected according to memory and processing resources of the end-device. Only parts of partitioned patches are decoded as a function of the selected sampling level.

Description

A METHOD AND APPARATUS FOR ENCODING AND DECODING VOLUMETRIC VIDEO AS PARTITIONED PATCH ATLASES

1. Technical Field

The present principles generally relate to the domain of three-dimensional (3D) scene and volumetric video content. The present document is also understood in the context of the encoding, the formatting and the decoding of data representative of the texture and the geometry of a 3D scene for a rendering of volumetric content on end-user devices such as mobile devices or Head- Mounted Displays (HMD).

2. Background

The present section is intended to introduce the reader to various aspects of art, which may be related to various aspects of the present principles that are described and/or claimed below. This discussion is believed to be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of the present principles. Accordingly, it should be understood that these statements are to be read in this light, and not as admissions of prior art.

The success of emerging technology like immersive media (MPEG Immersive Video - MIV based on a 3DoF+ approach) is often determined by the wide adoption of this technology on everyday devices like on embedded low-end user devices like digital TV, set-top-boxes or tablets, or wireless HMD. However, these devices have low GPU resource and the immersive technology must be also adapted for this type of devices. A possible approach could be to generate two types of content, one degraded immersive content for low-end user devices, and one to get richer immersive experience on high-end user gamers’ PC. Such an approach requires the storage of several versions of a same content and the knowledge of the type of client device for the transmission of the adapted version. Another approach is based on tiling methods. However, such an approach has the drawback to introduce latency in the streaming as the server has to be informed of the location and direction of every immersive rendering client devices to send the corresponding tiles. There is a lack of a data stream format for encoding volumetric video which enables a low- end decoder to provide a same level of experience than a high-end decoder from a same source.

3. Summary

The following presents a simplified summary of the present principles to provide a basic understanding of some aspects of the present principles. This summary is not an extensive overview of the present principles. It is not intended to identify key or critical elements of the present principles. The following summary merely presents some aspects of the present principles in a simplified form as a prelude to the more detailed description provided below.

In the context of a patch-based transmission of a volumetric video content, the present principles relate to a method implemented in a device comprising a memory and a processor. The method comprises:

- decoding, from a data stream, an atlas image packing patch pictures, projection metadata and partitioning metadata indicating a partition mode for each patch picture; a patch picture being a projection of a part of a 3D scene;

- re-arranging pixels of the patch pictures according to the associated partitioning metadata and according to memory and/or processing capabilities of the device; and

- generating a viewport image representative of the 3D scene from a viewpoint in the 3D scene based on there-arranged patch pictures and the projection metadata.

The present principles also relate to a method comprising:

- obtaining patch pictures and associated projection metadata; a patch picture being a projection of a part of a 3D scene;

- selecting a partition mode for each of the patch pictures;

- re-arranging pixels of the patch pictures according to the corresponding selected partition mode;

- encoding, in a data stream, the patch pictures packed in an atlas image, the associated projection metadata and partitioning metadata indicating the partition mode of each patch picture. The present principles also relate to a device comprising a processor configured to implement the two methods above and a data stream encoding and embedding data representative of said volumetric content, in particular metadata indicating whether a patch is partitioned and, if so, according to which partitioning mode.

4. Brief Description of Drawings

The present disclosure will be better understood, and other specific features and advantages will emerge upon reading the following description, the description making reference to the annexed drawings wherein:

- Figure 1 shows an example of an atlas encoding the texture information (e.g. RGB data or YUV data) of the points of a 3D scene, according to a non-limiting embodiment of the present principles;

- Figure 2 shows a non-limitative example of the encoding, transmission and decoding of data representative of a sequence of 3D scenes;

- Figure 3 shows an example architecture of a device which may be configured to implement a method described in relation with Figures 6 and 7;

- Figure 4 shows an example of an embodiment of the syntax of a stream when the data are transmitted over a packet-based transmission protocol;

- Figure 5 illustrates a new internal format for a patch belonging to an atlas according to a non-limiting embodiment of the present principles;

- Figure 6 illustrates a method of generating a patch according to the present principles;

- Figure 7 illustrates a method 70 for encoding a patch atlas representative of a volumetric content according to a partition mode.

5. Detailed description of embodiments

The present principles will be described more fully hereinafter with reference to the accompanying figures, in which examples of the present principles are shown. The present principles may, however, be embodied in many alternate forms and should not be construed as limited to the examples set forth herein. Accordingly, while the present principles are susceptible to various modifications and alternative forms, specific examples thereof are shown by way of examples in the drawings and will herein be described in detail. It should be understood, however, that there is no intent to limit the present principles to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present principles as defined by the claims.

The terminology used herein is for the purpose of describing particular examples only and is not intended to be limiting of the present principles. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises", "comprising," "includes" and/or "including" when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. Moreover, when an element is referred to as being "responsive" or "connected" to another element, it can be directly responsive or connected to the other element, or intervening elements may be present. In contrast, when an element is referred to as being "directly responsive" or "directly connected" to other element, there are no intervening elements present. As used herein the term "and/or" includes any and all combinations of one or more of the associated listed items and may be abbreviated as"/".

It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element without departing from the teachings of the present principles.

Although some of the diagrams include arrows on communication paths to show a primary direction of communication, it is to be understood that communication may occur in the opposite direction to the depicted arrows.

Some examples are described with regard to block diagrams and operational flowcharts in which each block represents a circuit element, module, or portion of code which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in other implementations, the function(s) noted in the blocks may occur out of the order noted. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending on the functionality involved.

Reference herein to “in accordance with an example” or “in an example” means that a particular feature, structure, or characteristic described in connection with the example can be included in at least one implementation of the present principles. The appearances of the phrase in accordance with an example” or “in an example” in various places in the specification are not necessarily all referring to the same example, nor are separate or alternative examples necessarily mutually exclusive of other examples.

Reference numerals appearing in the claims are by way of illustration only and shall have no limiting effect on the scope of the claims. While not explicitly described, the present examples and variants may be employed in any combination or sub-combination.

Immersive video, also called 360° flat video, allows the user to watch all around himself through rotations of his head around a still point of view. Rotations only allow a 3 Degrees of Freedom (3DoF) experience. Even if 3DoF video is sufficient for a first omnidirectional video experience, for example using a Head-Mounted Display device (HMD), 3DoF video may quickly become frustrating for the viewer who would expect more freedom, for example by experiencing parallax. In addition, 3DoF may also induce dizziness because of a user never only rotates his head but also translates his head in three directions, translations which are not reproduced in 3DoF video experiences.

A large field-of-view content may be, among others, a three-dimension computer graphic imagery scene (3D CGI scene), a point cloud or an immersive video. Many terms might be used to design such immersive videos: Virtual Reality (VR), 360, panoramic, 4p steradians, immersive, omnidirectional or large field of view for example.

Volumetric video (also known as 6 Degrees of Freedom (6DoF) video) is an alternative to 3DoF video. When watching a 6DoF video, in addition to rotations, the user can also translate his head, and even his body, within the watched content and experience parallax and even volumes. Such videos considerably increase the feeling of immersion and the perception of the scene depth and prevent from dizziness by providing consistent visual feedback during head translations. The content is created by the means of dedicated sensors allowing the simultaneous recording of color and depth of the scene of interest. The use of rig of color cameras combined with photogrammetry techniques is a way to perform such a recording, even if technical difficulties remain.

While 3DoF videos comprise a sequence of images resulting from the un-mapping of texture images (e.g. spherical images encoded according to latitude/longitude projection mapping or equirectangular projection mapping), 6DoF video frames embed information from several points of views. They can be viewed as a temporal series of point clouds resulting from a three- dimension capture. Two kinds of volumetric videos may be considered depending on the viewing conditions. A first one (i.e. complete 6DoF) allows a complete free navigation within the video content whereas a second one (aka. 3DoF+) restricts the user viewing space to a limited volume called viewing bounding box, allowing limited translation of the head and parallax experience. This second context is a valuable trade-off between free navigation and passive viewing conditions of a seated audience member.

A technical approach for the encoding of volumetric video is based on the projection of the 3D scene onto a multiplicity of 2D images, called patches, packed into atlases which can be further compressed using conventional video encoding standards (e.g., HEVC). At the decoding, using every patch of an atlas to prepare the rendering of the 3D scene in the current viewport is not always required, nor desirable. Indeed, some patches comprise information about points that are not visible from the current point of view and, for a given part of the 3D scene to be rendered, some patches comprise information more accurate than other patches comprising information for the same part of the scene when viewed from the current point of view.

Figure 1 shows an example of an atlas 10 encoding the texture information (e.g. RGB data or YUV data) of the points of a 3D scene, according to a non-limiting embodiment of the present principles. An atlas is an image packing patches, a patch being a picture obtained by projecting a part of the points of the 3D scene on a surface located in the space of the 3D scene. From a patch to another one, this surface may be located at different places within the space of the 3D scene and only a sample of the points of the scene (for instance, points not already projected onto another surface) is projected to generate a patch picture. The layout of an atlas is the way patches are organized on the image plane of the atlas. In the example of figure 1, atlas 10 comprises a first part 11 comprising the texture information of the points of the 3D scene that are visible from a point of view and one or second parts 12. The same atlas with the same layout is created to encode geometry information (i.e. pixels of the geometry atlas encoding the depth of projected points, that is the distance, in the Euclidian space of the 3D scene, between a given point and its projected counterpart onto the projection surface). In a variant, pixels of the atlas have four components (e.g. RGBA) and color and geometry are encoded in a single atlas. The texture information of the first part 11 may for example be obtained according to an equirectangular projection mapping, an equirectangular projection mapping being an example of spherical projection mapping. In the example of figure 1, the second parts 12 are arranged at the left and right borders of the first part 11. They may be arranged differently. Second parts 12 comprise texture information of parts of the 3D scene that are complementary to the part visible from the point of view. The second parts may be obtained by removing from the 3D scene the points that are visible from the first viewpoint (the texture of which being stored in the first part 11) and by projecting the remaining points according to the same point of view. The latter process may be reiterated iteratively to obtain at each time the hidden parts of the 3D scene. According to a variant, the second parts may be obtained by removing from the 3D scene the points that are visible from the point of view, for example a central point of view (the texture of which being stored in the first part) and by projecting the remaining points according to a point of view different from the central point of view, for example from one or more second point of view of a space of view centred onto the central point of view (e.g. the viewing space of a 3DoF rendering).

The first part 11 may be seen as a first large texture patch (corresponding to a first part of the 3D scene) and the second parts 12 comprises smaller textures patches (corresponding to second parts of the 3D scene that are complementary to the first part). Such an atlas has the advantage to be compatible at the same time with 3DoF rendering (when rendering only the first part 61) and with 3DoF+ / 6DoF rendering. Part 11 comprises a single big patch usual called central patch while parts 12 comprise a plurality of small patches usually called peripheral patches or parallax patches (because they are used to decode points of the scene visible from a non-central point of view and discovered thanks to the parallax effect). Figure 2 shows a non-limitative example of the encoding, transmission and decoding of data representative of a sequence of 3D scenes. The encoding format that may be, for example and at the same time, compatible for 3DoF, 3DoF+ and 6DoF decoding.

A sequence of 3D scenes 20 is obtained. As a sequence of pictures is a 2D video, a sequence of 3D scenes is a 3D (also called volumetric) video. A sequence of 3D scenes is provided according to a viewing zone from which the 3D scene may be viewed at the decoding side. 3D scenes may be provided to a volumetric video rendering device for a 3DoF, 3Dof+ or 6DoF rendering and displaying.

Sequence of 3D scenes 20 is provided to an encoder 21. The encoder 21 takes one 3D scenes or a sequence of 3D scenes as input and provides a bit stream representative of the input. The bit stream may be stored in a memory 22 and/or on an electronic data medium and may be transmitted over a network 22. The bit stream representative of a sequence of 3D scenes may be read from a memory 22 and/or received from a network 22 by a decoder 23. Decoder 23 is inputted by said bit stream and provides a sequence of 3D scenes, for instance in a point cloud format.

Encoder 21 may comprise several circuits implementing several steps. In a first step, encoder 21 projects each 3D scene onto at least one 2D picture. 3D projection is any method of mapping three-dimensional points to a two-dimensional plane. As most current methods for displaying graphical data are based on planar (pixel information from several bit planes) two- dimensional media, the use of this type of projection is widespread, especially in computer graphics, engineering and drafting. Projection circuit 211 provides at least one two-dimensional frame 2111 for a 3D scene of sequence 20. Frame 2111 comprises color information and depth information representative of the 3D scene projected onto frame 2111. In a variant, color information and depth information are encoded in two separate frames 2111 and 2112.

Metadata 212 are used and updated by projection circuit 211. Metadata 212 comprise information about the projection operation (e.g. projection parameters) and about the way color and depth information is organized within frames 2111 and 2112 as described in relation to figures 5 to 7.

According to the present principles, at least one validity domain is determined by projection circuit 211. A validity domain is an information describing a part of the viewing zone. A validity domain may be representative of a connected 3D region or a possibly disconnected union of connected regions. Projection circuit 211 associates every patch it generated with a validity domain. In a variant, projection circuit 211 does not associate some of the patches with any validity domain, indicating that the validity domain of this patch is the entire viewing zone. A part of the 3D space is delimited and associated with a patch when the information projected onto this patch are necessary and accurate to re-build the 3D scene for a view from a point of view encompassed in this part of the 3D space at the rendering. A same validity domain may be associated with several patches. A description of determined validity domains and the associations between patches and validity domains is added to metadata 212, so this information is going to be encapsulated in the data stream by data encapsulation circuit 214.

In an embodiment of the present principles, validity domains are determined in relation to centers of projection. The validity domain determined for one center of projection is associated with every patch generated according to this projection center as described in reference to figure 5. Descriptions of validity domains added in metadata 212 are associated with projection data describing the different projection operations used to generate patches of the atlas.

A video encoding circuit 213 encodes sequence of frames 2111 and 2112 as a video. Pictures of a 3D scene 2111 and 2112 (or a sequence of pictures of the 3D scene) is encoded in a stream by video encoder 213. Then video data and metadata 212 are encapsulated in a data stream by a data encapsulation circuit 214.

Encoder 213 is for example compliant with an encoder such as:

- JPEG, specification ISO/CEI 10918-1 UIT-T Recommendation T.81, https://www.itu.int/rec/T-REC-T.81/en;

- AVC, also named MPEG-4 AVC or h264. Specified in both UIT-T H.264 and ISO/CEI MPEG-4 Part 10 (ISO/CEI 14496-10), http://www.itu.int/rec/T-REC-H.264/en, HEVC (its specification is found at the ITU website, T recommendation, H series, h265, http://www.itu.int/rec/T-REC-H.265-201612-Een);

- 3D-HEVC (an extension of HEVC whose specification is found at the ITU website, T recommendation, H series, h265, http://www.itu.int/rec/T-REC-H.265-201612-I/en annex G and I); - VP9 developed by Google; or

- AVI (AOMedia Video 1) developed by Alliance for Open Media.

The data stream is stored in a memory that is accessible, for example through a network 22, by a decoder 23. Decoder 23 comprises different circuits implementing different steps of the decoding. Decoder 23 takes a data stream generated by an encoder 21 as an input and provides a sequence of 3D scenes 24 to be rendered and displayed by a volumetric video display device, like a Head-Mounted Device (HMD). Decoder 23 obtains the stream from a source 22. For example, source 22 belongs to a set comprising:

- a local memory, e.g. a video memory or a RAM (or Random-Access Memory), a flash memory, a ROM (or Read Only Memory), a hard disk;

- a storage interface, e.g. an interface with a mass storage, a RAM, a flash memory, a ROM, an optical disc or a magnetic support;

- a communication interface, e.g. a wireline interface (for example a bus interface, a wide area network interface, a local area network interface) or a wireless interface (such as a IEEE 802.11 interface or a Bluetooth® interface); and

- a user interface such as a Graphical User Interface enabling a user to input data.

Decoder 23 comprises a circuit 234 for extract data encoded in the data stream. Circuit 234 takes a data stream as input and provides metadata 232 corresponding to metadata 212 encoded in the stream and a two-dimensional video. The video is decoded by a video decoder 233 which provides a sequence of frames. Decoded frames comprise color and depth information. A frame is an atlas, that is image data comprising a set of patches packed in the frame. A patch is image data comprising information to retrieve points of the 3D scene to reconstruct. In a variant, video decoder 233 provides two sequences of frames, one comprising color information, the other comprising depth information.

According to the present principles, metadata 232 comprise at least one validity domain associated with at least one patch of the atlas. A validity domain is an information representative of a part of said viewing zone of the 3D space of the 3D scene. A circuit 231 uses metadata 232 to un-project color and depth information from decoded frames to provide a sequence of 3D scenes 24. Sequence of 3D scenes 24 corresponds to sequence of 3D scenes 20, with a possible loss of precision related to the encoding as a 2D video and to the video compression.

A 3D scene is retrieved from an atlas in which a plurality of patches is packed. According to the present principles, circuit 231 de-project pixels of a subset of the patches of the atlas. Circuit 231 selects only patches which are associated (via metadata 232) to a validity domain encompassing the current point of view of the rendering. In a variant, if a patch is associated to no validity domain, this patch is always used for de-projecting.

In an embodiment of the present principles, metadata 232 comprise a collection of a patch data. In a patch data item, the patch is associated with projection data comprising parameters of the projection operation used for generating this patch. Metadata 232 also comprise a collection of projection data and a projection data is associated with a validity domain. In this embodiment, circuit 231 selects, for de-projecting, the patches which are associated with a projection data which is, itself, associated with a validity domain encompassing the current point of view.

Figure 3 shows an example architecture of a device 30 which may be configured to implement a method described in relation with figures 6 and 7. Encoder 21 and/or decoder 23 of figure 2 may implement this architecture. Alternatively, each circuit of encoder 21 and/or decoder 23 may be a device according to the architecture of figure 3, linked together, for instance, via their bus 31 and/or via I/O interface 36.

Device 30 comprises following elements that are linked together by a data and address bus

31:

- a microprocessor 32 (or CPU), which is, for example, a DSP (or Digital Signal Processor);

- a ROM (or Read Only Memory) 33;

- a RAM (or Random Access Memory) 34;

- a storage interface 35;

- an I/O interface 36 for reception of data to transmit, from an application; and

- a power supply, e.g. a battery. In accordance with an example, the power supply is external to the device. In each of mentioned memory, the word « register » used in the specification may correspond to area of small capacity (some bits) or to very large area (e.g. a whole program or large amount of received or decoded data). The ROM 33 comprises at least a program and parameters. The ROM 33 may store algorithms and instructions to perform techniques in accordance with present principles. When switched on, the CPU 32 uploads the program in the RAM and executes the corresponding instructions.

The RAM 34 comprises, in a register, the program executed by the CPU 32 and uploaded after switch-on of the device 30, input data in a register, intermediate data in different states of the method in a register, and other variables used for the execution of the method in a register.

The implementations described herein may be implemented in, for example, a method or a process, an apparatus, a computer program product, a data stream, or a signal. Even if only discussed in the context of a single form of implementation (for example, discussed only as a method or a device), the implementation of features discussed may also be implemented in other forms (for example a program). An apparatus may be implemented in, for example, appropriate hardware, software, and firmware. The methods may be implemented in, for example, an apparatus such as, for example, a processor, which refers to processing devices in general, including, for example, a computer, a microprocessor, an integrated circuit, or a programmable logic device. Processors also include communication devices, such as, for example, computers, cell phones, portable/personal digital assistants ("PDAs"), and other devices that facilitate communication of information between end-users.

In accordance with examples, the device 30 is configured to implement a method described in relation to figure 6 and 7, and belongs to a set comprising:

- a mobile device;

- a communication device;

- a game device;

- a tablet (or tablet computer);

- a laptop;

- a still picture camera; - a video camera;

- an encoding chip;

- a server (e.g. a broadcast server, a video-on-demand server or a web server).

Figure 4 shows an example of an embodiment of the syntax of a stream when the data are transmitted over a packet-based transmission protocol. Figure 4 shows an example structure 4 of a volumetric video stream. The structure consists in a container which organizes the stream in independent elements of syntax. The structure may comprise a header part 41 which is a set of data common to every syntax elements of the stream. For example, the header part comprises some of metadata about syntax elements, describing the nature and the role of each of them. The header part may also comprise a part of metadata 212 of figure 2, for instance the coordinates of a central point of view used for projecting points of a 3D scene onto frames 2111 and 2112. The structure comprises a payload comprising an element of syntax 42 and at least one element of syntax 43. Syntax element 42 comprises data representative of the color and depth frames. Images may have been compressed according to a video compression method.

Element of syntax 43 is a part of the payload of the data stream and may comprise metadata about how frames of element of syntax 42 are encoded, for instance parameters used for projecting and packing points of a 3D scene onto frames. Such metadata may be associated with each frame of the video or to group of frames (also known as Group of Pictures (GoP) in video compression standards). According to the present principles, metadata of element of syntax 43 also comprise at least one validity domain associated with at least one patch of the atlas. A validity domain is an information representative of a part of said viewing zone of the 3D space of the 3D scene and may be encoded according to different representations and structures. Examples of such representations and structures are provided in the present disclosure.

For example, metadata comprise a collection of at least one validity domain. Items of this collection may be indexed, for example by a letter or by an integer. Metadata further comprise a collection of projection data comprising at least a reference to one of the validity domains. The reference may be, for instance the index of the validity domain in the collection or a pointer to the memory address of this data. In this example, metadata also comprise a collection of patch data comprising a reference to a patch and a reference to one of the projection data. The consumption of omnidirectional atlases is challenging on embedded devices. Ensuring a good visual quality, at a resolution of 15 pixels per degree (the usual standard on HMD), requires an atlas composed of more than 14M pixels per frame - 5.3Kx2.65K. Assuming that it has the decoding capabilities, a low-end device may be able to render such a frame, but at the cost of an increased rendering time, i.e. with a framerate below 30 FPS, inducing missing frames.

So, playing immersive volumetric content (MPEG Immersive Video - MIV based on a 3DoF+ approach) on a low-end device, like a wireless HMD (for example, the Oculus Quest) is a challenge as memory and processing resources may be limited to process high-resolutions source images. Such devices have lower capacity of parallel processing and lower memory bandwidth compared to high-end PC or consoles. Those characteristics are critical for rendering a volumetric content at a visually acceptable frame rate.

A possible approach is to generate two types of content, one lightweight immersive content for low-end devices and one to get richer immersive experience on high-end devices.

There is a lack for a single unified format for a volumetric content, addressing any type of devices with the following benefits:

• playable by a low-end device

• avoiding fragmentation and redundancy on server storage; a same content is used by both type of devices (no generation of different quality content files requiring expensive transcoding and storing operations).

• dynamic adaptation of the quality of the rendering by the target device.

Experience shows that that the time required for rendering a volumetric frame is correlated to the percentage of pixels read (i.e. accessed by the GPU) in the atlas. Rendering time is a linear function of the number of pixels accessed in the atlas. Thus, decreasing the number of memory lookups for an atlas of a given size is a major criterion to enhance the rendering speed on low-end devices. A problem addressed by the present principles is, then, how to decrease the number of pixels to be read, without penalizing the high-end devices and without requiring extra patches in lower resolution (than the atlas resolution), as it would introduce pixel redundancy and decoding costs.

A possible approach is to perform sparse read of pixels in the atlas, like reading every two pixels. However, such an approach does not solve the problem because the internal design of GPU caches is adapted to provide an efficient access to continuous lists of data and is losing in performance when a ‘jump’ has to be performed in the access to the memory stream. GPU caches are organized in 8-way banks memory system and each fetch of a pixel from a GPU texture also loads every 2D neighbor pixel into a 2D texture cache. Therefore, the number of memory stream reads in a scattered manner would be equivalent to the number of reads for the whole atlas. Additionally, reading sparse pixels in a texture image induces bad read performance. Scattered accesses in memory is equivalent to random access: it breaks cache locality, induces not coalesced reads into shared memory and increases the number of cache misses.

Figure 5 illustrates a new internal format for a patch belonging to an atlas according to a non-limiting embodiment of the present principles. A texture format for a 2D patch resulting from a projection into a view is proposed. Such a patch is associated with metadata for its signalization, that indicates to the decoder how to interpret the content and access subpart of it. The proposed internal format allows subsampling access of the patch content, to offer different levels of detail on heterogeneous devices. In an advantageous embodiment, this format is applied at least to the central patch of an immersive atlas as illustrated in Figure 1. A partitioned patch 52 is obtained by dividing the original patch 51 into a composition of Nhorizontai ^x Nvenicai subdivision patches. The sub-sampling process is a way of reorganizing the position of each pixel within a partitioned patch, so that the Tenderer can read a contiguous subset of all pixels of the patch, for a lower resolution (e.g. half resolution on Figure 5).

A frame of the content prepared according to this embodiment is sent as is to any end- device. Then, it can be rendered with different resolutions (four in the example of Figure 5) according to the end-device memory and processing capabilities. Thus, a low-end device will only parse one fourth of the decoded frame to render it with one fourth of the full resolution in a timely manner. A medium end-device may parse 2 or 3 sub-patches whereas a premium end-device may parse the whole composition of the four sub-patches to render it at full resolution. According to the present principles, atlases are encoded at a low encoding cost. A unified generic content is used for any type of devices, this generic content may be scaled with the GPU performance of any device, inducing no redundancy storage cost on server side.

Three entities of the delivery chain are involved by the present principles: the content ingest for the generation and the encoding of the content, the signaling between the server and the end- device and the decoder for the rendering step.

The process of generating an immersive format is modified to insert a partitioning step after the packing phase. All the patches are not necessarily partitioned as it is mainly interesting for the bigger ones, like the central 360° ERP patch of figure 1; whereas, it is possible to apply the partitioning process to a subset or to the whole set of patches.

For a source patch P to be partitioned into a target patch P’, of size W ^c H, the encoder:

• Obtains a number of partitioning for horizontal and vertical direction, Nhorizontai, Nvenicai, defines subdivisions areas Sij inside the target patch P’, for i in [0 - Nhorizontai- 1] and j in [0 - Nvenicai- 1 ] . In an embodiment, Sy is a rectangular area of left bottom origin Oy and dimensions Wy x Hy . Herein and without loss of generality, these rectangles are equi- dimensioned, e.g. of origin position Oy = (i x W/NhorizontaiJ ^x H/Nverticai) and dimensions Wij x Hij = W/ horizontal X H/Nverticalj

• Builds subsets of pixels by fetching horizontally and vertically interleaved pixels from the source patch, and put them packed together in one of the subdivisions Sy of P’;

• Generates the metadata to signal differently the partitioned patches.

Patches P and P’ have the same size. So, packing of patches inside the atlas is not altered. P replaced by P’ at the same packing position in the atlas. A patch P may be partitioned according different partitioned modes. A partition mode is a parametrized method to re-organize pixels of a patch. In the examples of figures 5 and 6 pixels of patch P are reorganized in sectors corresponding to their coordinates in patch P. Numbers Nhorizontai (called n in the claims) and Nvenicai (called m in the claims) ar the parameters of such a partition mode. Different methods may be used to partition a patch. For instance, pixels of a given sub-sampled sector may be arranged in rows or in columns. As the number of pixels does not vary, the size (i.e. width x height) of the partitioned patch may remain the same. In another embodiment, the partition mode may modify the shape of patch P. Patches are projections of a part of the 3D scene and, as so, they comprise unused pixels, that are pixels of the patch on which no point of the 3D scene has been projected. In the example of figure 6, empty squares are unused pixels. A partition of a patch also comprises unused pixels. A partition mode may take advantage of this fact by grouping useful pixels in a rectangle according to a mode, for example, based on the shape of the useful pixel area and/or the shape of the unused pixel area. For instance, partition C of patch 602 may be reshaped in a 2x2 square according to the recognition of the L shape of its useful pixels. Parameters of such a regrouping mode has to be added to the partition metadata of the patch. In such an embodiment, the partitioning step is advantageously performed before the packing step. Indeed, the packing step is performed once the shape and size of patches to be packed in the atlas is known.

An example correspondence between a source pixel, of relative coordinates (u,v) within P (starting at origin (0,0), direct orthonormal system) and its target position (upartitition,Vpartition) in the partitioned patch P’ may be defined by the following equations:

Figure 6 illustrates a method of generating a patch according to the present principles. A patch sub-composition is presented here for Nhorizontai = Nverticai= 2, corresponding to 4 sub-sampled subdivisions. At a step 61, an atlas comprising a patch 601 of dimensions W x H is obtained. In a step 62, a patch 602 is generated by organizing pixels of patch 601 as the union of four sub divisions Sij of dimensions W/2 x H/2, noted A, B, C and D on the example of Figure 6, where:

• So,i contains pixels with even x, odd y of P, noted A

• Si,i contains pixels with odd x, odd y of P, noted B · Si, o contains pixels with odd x, even y of P, noted C

• So,o contains pixels with even x, even y of P, noted D

At step 62, patch 601 is replaced in the atlas by patch 602. At a step 63, the atlas is encoded in association with metadata in a data stream. The atlas is, for example, is HEVC-encoded. To optimize compression and to avoid overhead, the Intra Block Copy (IBC) extension of HEVC may be set. This extension allows, for a given frame, to predict a macroblock from another one. Indeed, a partition of a partitioned patch may be efficiently predicted from the first one. A specialized encoder may take Nhorizontai and Nvenicai as parameters and be parametrized to take advantage of the specific structures of blocks of the atlas image according to the present principles.

The signalization metadata comprise three dedicated attributes: · patch_type: Enum field to indicate the patch type: {raw, partitioned}.

• number_of_ horizontal subdivisions: uint(8) field to specify Nhorizontai

• number_of_ vertical subdivisions: uint(8) field to specify Nvenicai

For example, these attributes may be included in metadata describing the patch list of the atlas in the stream metadata as proposed in the table below.

At decoder side, the end-device reads metadata information, for instance at each Group Of Pictures (GOP) interval, to build a patch map texture for an atlas (a). This data patchmap_2D [a] gives for each pixel (x,y) its patch identifier value patch_id. The patch identifier may be used later, for example by a vertex shader to fetch information related to this patch (e.g. position of the patch projection surface in the scene, position of the patch in the atlas, etc.).

Then in a shader, each pixel (x,y) read from an atlas (a) and belonging to a patch of id=patch_id is converted to its view position (x_view,yview). Pixels coordinates conversion from the atlas to the view is proposed in the following equations.

This conversion formula (x,y) to (x_view,yview) is adapted in the case of a partitioned patch.

A partitioned position is first computed. Pixel position (u_Partition,Vpartition) is relative to the patch position in the atlas and then deduced from it the subdivision indices (i,j) and the relative position (usubdivision, v subdivision) in the patch subdivision (A,B,C or D) to which it belongs (i is the horizontal index of the subdivision, j is the vertical index of the subdivision). From this relative subdivision position, the original (u,v) relative position of the pixel is computed, before than the patch P is partitioned. A conversion example of atlas coordinates to view coordinates according to one embodiment of the invention is provided by the method below:

To take benefit of the present principles, a rendering shader does not read all pixels of a partitioned patch, but only a subset of atlas pixels depending on the desired sampling level. Thus, for a sampling level of:

• 100 % of pixels: read all pixels in A, B, C, D

• 75% of pixels: read all pixels in A, B, C

• 50% of pixels: read all pixels in A, B

• 25% of pixels: read all pixels in A

An example implementation is to build a mesh of pixels at each new patch map layout (e.g. at each GOP), specific to the expected sampling level. By reading the patch list in metadata, a renderer can construct this mesh: the mesh is the union of all 2D atlas positions (x,y) for pixels belonging to each RAW patches and also belonging to the wanted subdivisions {A,B,C,D} of each partitioned patches. This mesh is used as input vertices for the shader described before.

Figure 7 illustrates a method 70 for encoding a patch atlas representative of a volumetric content according to a partition mode. At a step 71, in a first embodiment, a patch atlas representative of the volumetric content is obtained. In a second embodiment, a set of patches is obtained, not already packed in nan atlas image is obtained. At a step 72, a sub-set of patches is selected to be partitioned. Selection of patches may be based on characteristics of patches and/or on known characteristics of the decoding end-device. For instance, every patch with a number of pixels greater than a given threshold are to be partitioned. Any criterion bringing a technical effect may be used for selection step 72. At least one patch is selected to be partitioned. At a step 73, selected patches are partitioned, that is, their pixels are re-organized according to a partition mode as described upper. In the first embodiment, the size and the shape of a partitioned patch are the same than the patch in the obtained atlas which is replaced by its counterpart. In the second embodiment, the set of patches, non-modified and partitioned patches, is packed in an atlas image. At a step 74, generated atlas and metadata signaling which patch is partitioned and according to which partition mode are encoded in association in a data stream.

The implementations described herein may be implemented in, for example, a method or a process, an apparatus, a computer program product, a data stream, or a signal. Even if only discussed in the context of a single form of implementation (for example, discussed only as a method or a device), the implementation of features discussed may also be implemented in other forms (for example a program). An apparatus may be implemented in, for example, appropriate hardware, software, and firmware. The methods may be implemented in, for example, an apparatus such as, for example, a processor, which refers to processing devices in general, including, for example, a computer, a microprocessor, an integrated circuit, or a programmable logic device. Processors also include communication devices, such as, for example, Smartphones, tablets, computers, mobile phones, portable/personal digital assistants ("PDAs"), and other devices that facilitate communication of information between end-users.

Implementations of the various processes and features described herein may be embodied in a variety of different equipment or applications, particularly, for example, equipment or applications associated with data encoding, data decoding, view generation, texture processing, and other processing of images and related texture information and/or depth information. Examples of such equipment include an encoder, a decoder, a post-processor processing output from a decoder, a pre-processor providing input to an encoder, a video coder, a video decoder, a video codec, a web server, a set-top box, a laptop, a personal computer, a cell phone, a PDA, and other communication devices. As should be clear, the equipment may be mobile and even installed in a mobile vehicle.

Additionally, the methods may be implemented by instructions being performed by a processor, and such instructions (and/or data values produced by an implementation) may be stored on a processor-readable medium such as, for example, an integrated circuit, a software carrier or other storage device such as, for example, a hard disk, a compact diskette (“CD”), an optical disc (such as, for example, a DVD, often referred to as a digital versatile disc or a digital video disc), a random access memory (“RAM”), or a read-only memory (“ROM”). The instructions may form an application program tangibly embodied on a processor-readable medium. Instructions may be, for example, in hardware, firmware, software, or a combination. Instructions may be found in, for example, an operating system, a separate application, or a combination of the two. A processor may be characterized, therefore, as, for example, both a device configured to carry out a process and a device that includes a processor-readable medium (such as a storage device) having instructions for carrying out a process. Further, a processor-readable medium may store, in addition to or in lieu of instructions, data values produced by an implementation.

As will be evident to one of skill in the art, implementations may produce a variety of signals formatted to carry information that may be, for example, stored or transmitted. The information may include, for example, instructions for performing a method, or data produced by one of the described implementations. For example, a signal may be formatted to carry as data the rules for writing or reading the syntax of a described embodiment, or to carry as data the actual syntax-values written by a described embodiment. Such a signal may be formatted, for example, as an electromagnetic wave (for example, using a radio frequency portion of spectrum) or as a baseband signal. The formatting may include, for example, encoding a data stream and modulating a carrier with the encoded data stream. The information that the signal carries may be, for example, analog or digital information. The signal may be transmitted over a variety of different wired or wireless links, as is known. The signal may be stored on a processor-readable medium.

A number of implementations have been described. Nevertheless, it will be understood that various modifications may be made. For example, elements of different implementations may be combined, supplemented, modified, or removed to produce other implementations. Additionally, one of ordinary skill will understand that other structures and processes may be substituted for those disclosed and the resulting implementations will perform at least substantially the same function(s), in at least substantially the same way(s), to achieve at least substantially the same result(s) as the implementations disclosed. Accordingly, these and other implementations are contemplated by this application.

Claims

1. A method implemented in a device comprising a memory and a processor comprising:

2. The method of claim 1, wherein the re-arranging preserves the shape of patch pictures.

3. The method of claim 1, wherein the re-arranging reshapes the patch pictures, parameters of the reshaping being decoded from the partitioning metadata.

4. A method comprising:

- selecting a partition mode for each of the patch pictures;

- encoding, in a data stream, the patch pictures packed in an atlas image, the associated projection metadata and partitioning metadata indicating the partition mode of each patch picture.

5. The method of claim 4, wherein obtained patch pictures are packed in a first atlas image, wherein the re-arranging preserves the shape of patch pictures and wherein the first atlas image is the atlas image encoded in the data stream.

6. The method of claim 4, wherein the re-arranging reshapes the patch pictures, parameters of the reshaping being encoded in the partitioning metadata.

7. The method of one of claims 4 to 6, wherein selecting a partition mode for a patch picture depends on a size of the patch picture.

8. A device comprising a memory and a processor configured to:

- decode, from a data stream, an atlas image packing patch pictures, projection metadata and partitioning metadata indicating a partition mode for each patch picture; a patch picture being a projection of a part of a 3D scene;

- re-arranging pixels of the patch pictures according to the associated partitioning metadata and according to memory and processing capabilities of the device; and

- generate a viewport image representative of the 3D scene from a viewpoint in the 3D scene based on the re-arranged patch pictures and the projection metadata.

9. The device of claim 8, wherein the re-arranging preserves the shape of patch pictures.

10. The device of claim 8, wherein the re-arranging reshapes the patch pictures, parameters of the reshaping being decoded from the partitioning metadata.

11. A device comprising a processor configured to:

- obtain an patch pictures and associated projection metadata; a patch picture being a projection of a part of a 3D scene;

- select a partition mode for each of the patch pictures;

- re-arrange pixels of the patch pictures according to the corresponding selected partition mode;

- encode, in a data stream, the patch pictures packed in an atlas image, the associated projection metadata and partitioning metadata indicating the partition mode of each patch picture.

12. The device of claim 11, wherein obtained patch pictures are packed in a first atlas image, wherein the re-arranging preserves the shape of patch pictures and wherein the first atlas image is the atlas image encoded in the data stream.

13. The device of claim 11, wherein the re-arranging reshapes the patch pictures, parameters of the reshaping being encoded in the partitioning metadata.

14. The device of one of claims 11 to 13, wherein selecting a partition mode for a patch picture depends on a size of the patch picture.

15. A data stream carrying data representative of a volumetric content, the data comprising an atlas image packing patch pictures and, for each patch picture, projection metadata and partitioning metadata indicating a partition mode; a patch picture being a projection of a part of a 3D scene.