Normalizing flow small architectures to code point cloud attributes

Abstract

Some embodiments of a method may include: obtaining a point cloud; performing feature enhancement on the point cloud; generating a latent space based on the information corresponding to the point cloud, wherein generating the data corresponding to the latent space includes: performing a subprocess twice, wherein the subprocess includes: passing the input data through a voxel shuffling layer to generate a voxel shuffling layer output; performing a convolution on the voxel shuffling layer output to generate convolution output data; passing the convolution output data through one or more coupling layers to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and passing the second loop subprocess output through an attention layer to generate the data corresponding to the latent space; and encoding the data corresponding to the latent space as a bitstream.

Description

NORMALIZING FLOW SMALL ARCHITECTURES TO CODE POINT CLOUD ATTRIBUTESCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims benefit of European Patent Application No. EP23306306, entitled "NORMALIZING FLOW SMALL ARCHITECTURES TO CODE POINT CLOUD ATTRIBUTES” and filed July 28, 2023, which is hereby incorporated by reference in its entirety.CROSS-REFERENCE TO OTHER APPLICATIONS

[0002] The present application incorporates by reference in their entirety the following applications: European Patent Application Serial No. EP22306317 entitled "METHODS AND APPARATUSES FOR ENCODING AND DECODING A POINT CLOUD” and filed September 6, 2022 ("‘317 application”); and European Patent Application Serial No. EP23305352, entitled "METHODS AND APPARATUSES FOR ENCODING AND DECODING A POINT CLOUD” and filed March 13, 2023 ("‘352 application”).BACKGROUND

[0003] The use of 3D applications is becoming more popular every day. To be able to exploit such applications, different data formats are being used. One of the data formats is point clouds. Point clouds are a set of unordered points with coordinates (x, y, z) corresponding to a point location in space and its attributes (such as colors and normal vectors).SUMMARY

[0004] Embodiments described herein include methods that are used in video encoding and decoding (collectively "coding”).

[0005] An example method in accordance with some embodiments may include: obtaining information corresponding to a point cloud; performing feature enhancement on point cloud data corresponding to the information; generating data corresponding to a latent space based on the information corresponding to the point cloud, wherein generating the data corresponding to the latent space may include: using the feature enhanced point cloud data as input data to a first loop through a subprocess; performing the subprocess twice, wherein the subprocess may include: passing the input data through a voxel shuffling layer to generatea voxel shuffling layer output; performing a convolution on the voxel shuffling layer output to generate convolution output data; passing the convolution output data through one or more coupling layers to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and passing the second loop subprocess output through an attention layer to generate the data corresponding to the latent space; and encoding the data corresponding to the latent space as a bitstream.

[0006] For some embodiments of the example method, the subprocess may be performed three times:

[0007] For some embodiments of the example method, the subprocess further includes performing an average back projection process on an output of the one or more coupling layers.

[0008] For some embodiments of the example method, performing the average back projection process may include performing at least one of an averaging process, a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

[0009] For some embodiments of the example method, the information corresponding to the point cloud further corresponds to one or more voxels.

[0010] For some embodiments of the example method, the subprocess is a normalizing flow (NF) subprocess.

[0011] For some embodiments of the example method, the encoded bitstream may include a signaling flag to indicate an architecture configuration.

[0012] An example apparatus in accordance with some embodiments may include: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform any of the methods listed above.

[0013] A second example method in accordance with some embodiments may include: obtaining information corresponding to a point cloud; generating data corresponding to a latent space based on the information corresponding to the point cloud, wherein generating the data corresponding to the latent space may include: using the feature enhanced point cloud data as input data to a first loop through a subprocess; performing the subprocess twice, wherein the subprocess may include: passing the input data through a voxel shuffling layer to generate a voxel shuffling layer output; performing a convolution on the voxel shuffling layer output to generate convolution output data; passing the convolution output data through one or more coupling layers to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and generating the data corresponding to the latent space based on the second loop subprocess output; and encoding the data corresponding to the latent space as a bitstream.

[0014] For some embodiments of the second example method, the subprocess is performed three times:

[0015] For some embodiments of the second example method, the subprocess further includes performing an average back projection process on an output of the one or more coupling layers.

[0016] For some embodiments of the second example method, performing the average back projection process may include performing at least one of an averaging process, a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

[0017] For some embodiments of the second example method, the information corresponding to the point cloud further corresponds to one or more voxels.

[0018] For some embodiments of the second example method, the subprocess is a normalizing flow (NF) subprocess.

[0019] For some embodiments of the second example method, the encoded bitstream may include a signaling flag to indicate an architecture configuration.

[0020] A second example apparatus in accordance with some embodiments may include: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform any of the methods listed above.

[0021] A third example method in accordance with some embodiments may include: obtaining an encoded bitstream; decoding the encoded bitstream to generate a latent space; and reconstructing a point cloud from the generated latent space, wherein reconstructing the point cloud from the generated latent space may include: passing the generated latent space through an attention layer to generate input data to a subprocess; performing the subprocess twice, wherein the subprocess may include: passing the input data through one or more coupling layers to generate a coupling layer output; performing a convolution on the coupling layer output to generate convolution output data; passing the convolution output data through a voxel shuffling layer to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and performing feature enhancement on the second loop subprocess output to generate the reconstructed point cloud.

[0022] For some embodiments of the third example method, the subprocess may be performed three times:

[0023] For some embodiments of the third example method, the subprocess further may include performing a copy back projection process on an output of the voxel shuffling layer.

[0024] For some embodiments of the third example method, performing the copy back projection process may include performing at least one of a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

[0025] For some embodiments of the third example method, the information corresponding to the point cloud further corresponds to one or more voxels.

[0026] For some embodiments of the third example method, the subprocess is a normalizing flow (NF) subprocess.

[0027] For some embodiments of the third example method, the encoded bitstream may include a signaling flag to indicate an architecture configuration.

[0028] A third example apparatus in accordance with some embodiments may include: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform any of the methods listed above.

[0029] A fourth example method / apparatus in accordance with some embodiments may include: obtaining an encoded bitstream; decoding the encoded bitstream to generate a latent space; and reconstructing a point cloud from the generated latent space, wherein reconstructing the point cloud from the generated latent space may include: obtaining input data to a subprocess based on the generated latent space; performing the subprocess twice, wherein the subprocess may include: passing the input data through one or more coupling layers to generate a coupling layer output; performing a convolution on the coupling layer output to generate convolution output data; passing the convolution output data through a voxel shuffling layer to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and generating the reconstructed point cloud based on the second loop subprocess output.

[0030] For some embodiments of the fourth example method, the subprocess may be performed three times:

[0031] For some embodiments of the fourth example method, the subprocess further may include performing a copy back projection process on an output of the voxel shuffling layer.

[0032] For some embodiments of the fourth example method, performing the copy back projection process includes performing at least one of a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

[0033] For some embodiments of the fourth example method, the information corresponding to the point cloud further corresponds to one or more voxels.

[0034] For some embodiments of the fourth example method, the subprocess is a normalizing flow (NF) subprocess.

[0035] For some embodiments of the fourth example method, the encoded bitstream may include a signaling flag to indicate an architecture configuration.

[0036] A fourth example apparatus in accordance with some embodiments may include: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform any of the methods listed above.

[0037] Another example apparatus in accordance with some embodiments may include at least one processor configured to perform one of the methods listed above.

[0038] Another example apparatus in accordance with some embodiments may include a computer- readable medium storing instructions for causing one or more processors to perform one of the methods listed above.

[0039] Another example apparatus in accordance with some embodiments may include at least one processor and at least one non-transitory computer-readable medium storing instructions for causing the at least one processor to perform any one of the methods listed above.

[0040] An example computer-readable medium in accordance with some embodiments may include storing a scene description file generated according to one of the methods listed above.

[0041] An example signal in accordance with some embodiments may include a scene description file generated according to any one of the methods listed above.

[0042] In additional embodiments, encoder and decoder apparatus are provided to perform the methods described herein. An encoder or decoder apparatus may include a processor configured to perform the methods described herein. The apparatus may include a computer-readable medium (e.g. a non-transitory medium) storing instructions for performing the methods described herein. In some embodiments, a computer-readable medium (e.g. a non-transitory medium) stores a video encoded using any of the methods described herein.

[0043] One or more of the present embodiments also provide a computer readable storage medium having stored thereon instructions for performing bi-directional optical flow, encoding or decoding video data according to any of the methods described above. The present embodiments also provide a computer readable storage medium having stored thereon a bitstream generated according to the methods described above. The present embodiments also provide a method and apparatus for transmitting the bitstreamgenerated according to the methods described above. The present embodiments also provide a computer program product including instructions for performing any of the methods described.BRIEF DESCRIPTION OF THE DRAWINGS

[0044] FIG. 1A is a system diagram illustrating an example communications system according to some embodiments.

[0045] FIG. 1 B is a system diagram illustrating an example wireless transmit / receive unit (WTRU) that may be used within the communications system illustrated in FIG. 1A according to some embodiments.

[0046] FIG. 1 C is a system diagram illustrating an example set of interfaces for a system according to some embodiments.

[0047] FIG. 2A is a block diagram showing an example system according to some embodiments.

[0048] FIG. 2B is a system diagram illustrating an example set of interfaces for networking according to some embodiments.

[0049] FIG. 2C is a message bitfield diagram showing an example packet according to some embodiments.

[0050] FIG. 3 is a process diagram showing an example normalizing flow architecture to compress and uncompress point cloud data according to some embodiments.

[0051] FIG. 4A is a process diagram illustrating an example feature enhancement layer according to some embodiments.

[0052] FIG. 4B is a process diagram illustrating an example feature enhancement according to some embodiments.

[0053] FIG. 4C is a process diagram illustrating an example coupling layer according to some embodiments.

[0054] FIG. 4D is a process diagram illustrating an example transformation block for coupling layers according to some embodiments.

[0055] FIG. 5 is a flowchart illustrating an example process for encoding a point cloud according to some embodiments.

[0056] FIG. 6 is a flowchart illustrating an example process for reconstructing a point cloud according to some embodiments.

[0057] FIG. 7 is a process diagram showing an example normalizing flow, non-averaged process according to some embodiments.

[0058] FIGs. 8A is a process diagram showing an example average back projection process according to some embodiments.

[0059] FIGs. 8B is a process diagram showing an example copy back projection process according to some embodiments.

[0060] FIG. 9 is a process diagram showing an example normalizing flow back projection process according to some embodiments.

[0061] FIG. 10 is a graph illustrating an example peak signal-to-noise ratio (PSNR) vs. bits per input point for a first test scenario according to some embodiments.

[0062] FIG. 11 is a graph illustrating an example peak signal-to-noise ratio (PSNR) vs. bits per input point for a second test scenario according to some embodiments.

[0063] FIG. 12 is a graph illustrating an example peak signal-to-noise ratio (PSNR) vs. bits per input point for a third test scenario according to some embodiments.

[0064] FIG. 13 is a flowchart illustrating an example process for encoding a point cloud according to some embodiments.

[0065] FIG. 14 is a flowchart illustrating an example process for decoding a point cloud according to some embodiments.

[0066] The entities, connections, arrangements, and the like that are depicted in— and described in connection with— the various figures are presented by way of example and not by way of limitation. As such, any and all statements or other indications as to what a particular figure "depicts,” what a particular element or entity in a particular figure "is” or "has,” and any and all similar statements— that may in isolation and out of context be read as absolute and therefore limiting— may only properly be read as being constructively preceded by a clause such as "In at least one embodiment, ... " For brevity and clarity of presentation, this implied leading clause is not repeated ad nauseum in the detailed description.DETAILED DESCRIPTION

[0067] FIG. 1A is a diagram illustrating an example communications system 100 in which one or more disclosed embodiments may be implemented. The communications system 100 may be a multiple access system that provides content, such as voice, data, video, messaging, broadcast, etc., to multiple wireless users. The communications system 100 may enable multiple wireless users to access such content throughthe sharing of system resources, including wireless bandwidth. For example, the communications systems 100 may employ one or more channel access methods, such as code division multiple access (CDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal FDMA (OFDMA), single-carrier FDMA (SC-FDMA), zero-tail unique-word DFT-Spread OFDM (ZT UW DTS-s OFDM), unique word OFDM (UW-OFDM), resource block-filtered OFDM, filter bank multicarrier (FBMC), and the like.

[0068] As shown in FIG. 1 A, the communications system 100 may include wireless transmit / receive units (WTRUs) 102a, 102b, 102c, 102d, a RAN 104 / 113, a ON 106, a public switched telephone network (PSTN) 108, the Internet 110, and other networks 112, though it will be appreciated that the disclosed embodiments contemplate any number of WTRUs, base stations, networks, and / or network elements. Each of the WTRUs 102a, 102b, 102c, 102d may be any type of device configured to operate and / or communicate in a wireless environment. By way of example, the WTRUs 102a, 102b, 102c, 102d, any of which may be referred to as a "station” and / or a "STA”, may be configured to transmit and / or receive wireless signals and may include a user equipment (UE), a mobile station, a fixed or mobile subscriber unit, a subscription-based unit, a pager, a cellular telephone, a personal digital assistant (PDA), a smartphone, a laptop, a netbook, a personal computer, a wireless sensor, a hotspot or Mi-Fl device, an Internet of Things (loT) device, a watch or other wearable, a head-mounted display (HMD), a vehicle, a drone, a medical device and applications (e.g., remote surgery), an industrial device and applications (e.g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain contexts), a consumer electronics device, a device operating on commercial and / or industrial wireless networks, and the like. Any of the WTRUs 102a, 102b, 102c and 102d may be interchangeably referred to as a UE.

[0069] The communications systems 100 may also include a base station 114a and / or a base station 114b. Each of the base stations 114a, 114b may be any type of device configured to wirelessly interface with at least one of the WTRUs 102a, 102b, 102c, 102d to facilitate access to one or more communication networks, such as the CN 106, the Internet 110, and / or the other networks 112. By way of example, the base stations 114a, 114b may be a base transceiver station (BTS), a Node-B, an eNode B, a Home Node B, a Home eNode B, a gNB, a NR NodeB, a site controller, an access point (AP), a wireless router, and the like. While the base stations 114a, 114b are each depicted as a single element, it will be appreciated that the base stations 114a, 114b may include any number of interconnected base stations and / or network elements.

[0070] The base station 114a may be part of the RAN 104 / 113, which may also include other base stations and / or network elements (not shown), such as a base station controller (BSC), a radio network controller (RNC), relay nodes, etc. The base station 114a and / or the base station 114b may be configured to transmit and / or receive wireless signals on one or more carrier frequencies, which may be referred to as a cell (notshown). These frequencies may be in licensed spectrum, unlicensed spectrum, or a combination of licensed and unlicensed spectrum. A cell may provide coverage for a wireless service to a specific geographical area that may be relatively fixed or that may change over time. The cell may further be divided into cell sectors. For example, the cell associated with the base station 114a may be divided into three sectors. Thus, in one embodiment, the base station 114a may include three transceivers, i.e., one for each sector of the cell. In an embodiment, the base station 114a may employ multiple-input multiple output (MIMO) technology and may utilize multiple transceivers for each sector of the cell. For example, beamforming may be used to transmit and / or receive signals in desired spatial directions.

[0071] The base stations 114a, 114b may communicate with one or more of the WTRUs 102a, 102b, 102c, 102d over an air interface 116, which may be any suitable wireless communication link (e.g., radio frequency (RF), microwave, centimeter wave, micrometer wave, infrared (IR), ultraviolet (UV), visible light, etc.). The air interface 116 may be established using any suitable radio access technology (RAT).

[0072] More specifically, as noted above, the communications system 100 may be a multiple access system and may employ one or more channel access schemes, such as CDMA, TDMA, FDMA, OFDMA, SC-FDMA, and the like. For example, the base station 114a in the RAN 104 / 113 and the WTRUs 102a, 102b, 102c may implement a radio technology such as Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access (UTRA), which may establish the air interface 116 using wideband CDMA (WCDMA). WCDMA may include communication protocols such as High-Speed Packet Access (HSPA) and / or Evolved HSPA (HSPA+). HSPA may include High-Speed Downlink (DL) Packet Access (HSDPA) and / or High-Speed UL Packet Access (HSUPA).

[0073] In an embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement a radio technology such as Evolved UMTS Terrestrial Radio Access (E-UTRA), which may establish the air interface 116 using Long Term Evolution (LTE) and / or LTE-Advanced (LTE-A) and / or LTE-Advanced Pro (LTE-A Pro).

[0074] In an embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement a radio technology such as NR Radio Access , which may establish the air interface 116 using New Radio (NR).

[0075] In an embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement multiple radio access technologies. For example, the base station 114a and the WTRUs 102a, 102b, 102c may implement LTE radio access and NR radio access together, for instance using dual connectivity (DC) principles. Thus, the air interface utilized by WTRUs 102a, 102b, 102c may be characterized by multipletypes of radio access technologies and / or transmissions sent to / from multiple types of base stations (e.g., a eNB and a gNB).

[0076] In other embodiments, the base station 114a and the WTRUs 102a, 102b, 102c may implement radio technologies such as IEEE 802.11 (i.e., Wireless Fidelity (WiFi), IEEE 802.16 (i.e., Worldwide Interoperability for Microwave Access (WiMAX)), CDMA2000, CDMA2000 1X, CDMA2000 EV-DO, Interim Standard 2000 (IS-2000), Interim Standard 95 (IS-95), Interim Standard 856 (IS-856), Global System for Mobile communications (GSM), Enhanced Data rates for GSM Evolution (EDGE), GSM EDGE (GERAN), and the like.

[0077] The base station 114b in FIG. 1A may be a wireless router, Home Node B, Home eNode B, or access point, for example, and may utilize any suitable RAT for facilitating wireless connectivity in a localized area, such as a place of business, a home, a vehicle, a campus, an industrial facility, an air corridor (e.g., for use by drones), a roadway, and the like. In one embodiment, the base station 114b and the WTRUs 102c, 102d may implement a radio technology such as IEEE 802.11 to establish a wireless local area network (WLAN). In an embodiment, the base station 114b and the WTRUs 102c, 102d may implement a radio technology such as IEEE 802.15 to establish a wireless personal area network (WPAN). In yet another embodiment, the base station 114b and the WTRUs 102c, 102d may utilize a cellular-based RAT (e.g., WCDMA, CDMA2000, GSM, LTE, LTE-A, LTE-A Pro, NR etc.) to establish a picocell or femtocell. As shown in FIG. 1A, the base station 114b may have a direct connection to the Internet 110. Thus, the base station 114b may not be required to access the Internet 110 via the CN 106.

[0078] The RAN 104 / 113 may be in communication with the CN 106, which may be any type of network configured to provide voice, data, applications, and / or voice over internet protocol (VoIP) services to one or more of the WTRUs 102a, 102b, 102c, 102d. The data may have varying quality of service (QoS) requirements, such as differing throughput requirements, latency requirements, error tolerance requirements, reliability requirements, data throughput requirements, mobility requirements, and the like. The CN 106 may provide call control, billing services, mobile location-based services, pre-paid calling, Internet connectivity, video distribution, etc., and / or perform high-level security functions, such as user authentication. Although not shown in FIG. 1A, it will be appreciated that the RAN 104 / 113 and / or the CN 106 may be in direct or indirect communication with other RANs that employ the same RAT as the RAN 104 / 113 or a different RAT. For example, in addition to being connected to the RAN 104 / 113, which may be utilizing a NR radio technology, the CN 106 may also be in communication with another RAN (not shown) employing a GSM, UMTS, CDMA 2000, WiMAX, E-UTRA, or WiFi radio technology.

[0079] The CN 106 may also serve as a gateway for the WTRUs 102a, 102b, 102c, 102d to access the PSTN 108, the Internet 110, and / or the other networks 112. The PSTN 108 may include circuit-switched telephone networks that provide plain old telephone service (POTS). The Internet 110 may include a global system of interconnected computer networks and devices that use common communication protocols, such as the transmission control protocol (TCP), user datagram protocol (UDP) and / or the internet protocol (IP) in the TCP / IP internet protocol suite. The networks 112 may include wired and / or wireless communications networks owned and / or operated by other service providers. For example, the networks 112 may include another CN connected to one or more RANs, which may employ the same RAT as the RAN 104 / 113 or a different RAT.

[0080] Some or all of the WTRUs 102a, 102b, 102c, 102d in the communications system 100 may include multi-mode capabilities (e.g., the WTRUs 102a, 102b, 102c, 102d may include multiple transceivers for communicating with different wireless networks over different wireless links). For example, the WTRU 102c shown in FIG. 1A may be configured to communicate with the base station 114a, which may employ a cellular-based radio technology, and with the base station 114b, which may employ an IEEE 802 radio technology.

[0081] FIG. 1 B is a system diagram illustrating an example WTRU 102. As shown in FIG. 1 B, the WTRU 102 may include a processor 118, a transceiver 120, a transmit / receive element 122, a speaker / microphone 124, a keypad 126, a display / touchpad 128, non-removable memory 130, removable memory 132, a power source 134, a global positioning system (GPS) chipset 136, and / or other peripherals 138, among others. It will be appreciated that the WTRU 102 may include any sub-combination of the foregoing elements while remaining consistent with an embodiment.

[0082] The processor 118 may be a general purpose processor, a special purpose processor, a conventional processor, a digital signal processor (DSP), a plurality of microprocessors, one or more microprocessors in association with a DSP core, a controller, a microcontroller, Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) circuits, any other type of integrated circuit (IC), a state machine, and the like. The processor 118 may perform signal coding, data processing, power control, input / output processing, and / or any other functionality that enables the WTRU 102 to operate in a wireless environment. The processor 118 may be coupled to the transceiver 120, which may be coupled to the transmit / receive element 122. While FIG. 1 B depicts the processor 118 and the transceiver 120 as separate components, it will be appreciated that the processor 118 and the transceiver 120 may be integrated together in an electronic package or chip.

[0083] The transmit / receive element 122 may be configured to transmit signals to, or receive signals from, a base station (e.g., the base station 114a) over the air interface 116. For example, in one embodiment, the transmit / receive element 122 may be an antenna configured to transmit and / or receive RF signals. In an embodiment, the transmit / receive element 122 may be an emitter / detector configured to transmit and / or receive IR, UV, or visible light signals, for example. In yet another embodiment, the transmit / receive element 122 may be configured to transmit and / or receive both RF and light signals. It will be appreciated that the transmit / receive element 122 may be configured to transmit and / or receive any combination of wireless signals.

[0084] Although the transmit / receive element 122 is depicted in FIG. 1 B as a single element, the WTRU 102 may include any number of transmit / receive elements 122. More specifically, the WTRU 102 may employ MIMO technology. Thus, in one embodiment, the WTRU 102 may include two or more transmit / receive elements 122 (e.g., multiple antennas) for transmitting and receiving wireless signals over the air interface 116.

[0085] The transceiver 120 may be configured to modulate the signals that are to be transmitted by the transmit / receive element 122 and to demodulate the signals that are received by the transmit / receive element 122. As noted above, the WTRU 102 may have multi-mode capabilities. Thus, the transceiver 120 may include multiple transceivers for enabling the WTRU 102 to communicate via multiple RATs, such as NR and IEEE 802.11 , for example.

[0086] The processor 118 of the WTRU 102 may be coupled to, and may receive user input data from, the speaker / microphone 124, the keypad 126, and / or the display / touchpad 128 (e.g., a liquid crystal display (LCD) display unit or organic light-emitting diode (OLED) display unit). The processor 118 may also output user data to the speaker / microphone 124, the keypad 126, and / or the display / touchpad 128. In addition, the processor 118 may access information from, and store data in, any type of suitable memory, such as the non-removable memory 130 and / or the removable memory 132. The non-removable memory 130 may include random-access memory (RAM), read-only memory (ROM), a hard disk, or any other type of memory storage device. The removable memory 132 may include a subscriber identity module (SIM) card, a memory stick, a secure digital (SD) memory card, and the like. In other embodiments, the processor 118 may access information from, and store data in, memory that is not physically located on the WTRU 102, such as on a server or a home computer (not shown).

[0087] The processor 118 may receive power from the power source 134, and may be configured to distribute and / or control the power to the other components in the WTRU 102. The power source 134 may be any suitable device for powering the WTRU 102. For example, the power source 134 may include one ormore dry cell batteries (e.g., nickel-cadmium (NiCd), nickel-zinc (NiZn), nickel metal hydride (NiMH), lithium- ion (Li-ion), etc.), solar cells, fuel cells, and the like.

[0088] The processor 118 may also be coupled to the GPS chipset 136, which may be configured to provide location information (e.g., longitude and latitude) regarding the current location of the WTRU 102. In addition to, or in lieu of, the information from the GPS chipset 136, the WTRU 102 may receive location information over the air interface 116 from a base station (e.g., base stations 114a, 114b) and / or determine its location based on the timing of the signals being received from two or more nearby base stations. It will be appreciated that the WTRU 102 may acquire location information by way of any suitable locationdetermination method while remaining consistent with an embodiment.

[0089] The processor 118 may further be coupled to other peripherals 138, which may include one or more software and / or hardware modules that provide additional features, functionality and / or wired or wireless connectivity. For example, the peripherals 138 may include an accelerometer, an e-compass, a satellite transceiver, a digital camera (for photographs and / or video), a universal serial bus (USB) port, a vibration device, a television transceiver, a hands free headset, a Bluetooth® module, a frequency modulated (FM) radio unit, a digital music player, a media player, a video game player module, an Internet browser, a Virtual Reality and / or Augmented Reality (VR / AR) device, an activity tracker, and the like. The peripherals 138 may include one or more sensors, the sensors may be one or more of a gyroscope, an accelerometer, a hall effect sensor, a magnetometer, an orientation sensor, a proximity sensor, a temperature sensor, a time sensor; a geolocation sensor; an altimeter, a light sensor, a touch sensor, a magnetometer, a barometer, a gesture sensor, a biometric sensor, and / or a humidity sensor.

[0090] The WTRU 102 may include a full duplex radio for which transmission and reception of some or all of the signals (e.g., associated with particular subframes for both the UL (e.g., for transmission) and downlink (e.g., for reception) may be concurrent and / or simultaneous. The full duplex radio may include an interference management unit to reduce and or substantially eliminate self-interference via either hardware (e.g., a choke) or signal processing via a processor (e.g., a separate processor (not shown) or via processor 118). In an embodiment, the WTRU 102 may include a half-duplex radio for which transmission and reception of some or all of the signals (e.g., associated with particular subframes for either the UL (e.g., for transmission) or the downlink (e.g., for reception)).

[0091] Although the WTRU is described in FIGs. 1A-1 B as a wireless terminal, it is contemplated that in certain representative embodiments that such a terminal may use (e.g., temporarily or permanently) wired communication interfaces with the communication network.

[0092] In representative embodiments, the other network 112 may be a WLAN.

[0093] In view of FIGs. 1A-1 B, and the corresponding description, one or more, or all, of the functions described herein may be performed by one or more emulation devices (not shown). The emulation devices may be one or more devices configured to emulate one or more, or all, of the functions described herein. For example, the emulation devices may be used to test other devices and / or to simulate network and / or WTRU functions.

[0094] The emulation devices may be designed to implement one or more tests of other devices in a lab environment and / or in an operator network environment. For example, the one or more emulation devices may perform the one or more, or all, functions while being fully or partially implemented and / or deployed as part of a wired and / or wireless communication network in order to test other devices within the communication network. The one or more emulation devices may perform the one or more, or all, functions while being temporarily implemented / deployed as part of a wired and / or wireless communication network. The emulation device may be directly coupled to another device for purposes of testing and / or may performing testing using over-the-air wireless communications.

[0095] The one or more emulation devices may perform the one or more, including all, functions while not being implemented / deployed as part of a wired and / or wireless communication network. For example, the emulation devices may be utilized in a testing scenario in a testing laboratory and / or a non-deployed (e.g., testing) wired and / or wireless communication network in order to implement testing of one or more components. The one or more emulation devices may be test equipment. Direct RF coupling and / or wireless communications via RF circuitry (e.g., which may include one or more antennas) may be used by the emulation devices to transmit and / or receive data.

[0096] FIG. 1 C is a system diagram illustrating an example set of interfaces for a system according to some embodiments. An extended reality display device, together with its control electronics, may be implemented using a system such as the system of FIG. 1 D. System 150 can be embodied as a device including the various components described below and is configured to perform one or more of the aspects described in this document. Examples of such devices, include, but are not limited to, various electronic devices such as personal computers, laptop computers, smartphones, tablet computers, digital multimedia set top boxes, digital television receivers, personal video recording systems, connected home appliances, and servers. Elements of system 150, singly or in combination, can be embodied in a single integrated circuit (IC), multiple ICs, and / or discrete components. For example, in at least one embodiment, the processing and encoder / decoder elements of system 150 are distributed across multiple ICs and / or discrete components. In various embodiments, the system 150 is communicatively coupled to one or more other systems, or other electronic devices, via, for example, a communications bus or through dedicated input and / or output ports.In various embodiments, the system 150 is configured to implement one or more of the aspects described in this document.

[0097] The system 150 includes at least one processor 152 configured to execute instructions loaded therein for implementing, for example, the various aspects described in this document. Processor 152 may include embedded memory, input output interface, and various other circuitries as known in the art. The system 150 includes at least one memory 154 (e.g., a volatile memory device, and / or a non-volatile memory device). System 150 may include a storage device 158, which can include non-volatile memory and / or volatile memory, including, but not limited to, Electrically Erasable Programmable Read-Only Memory (EEPROM), Read-Only Memory (ROM), Programmable Read-Only Memory (PROM), Random Access Memory (RAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), flash, magnetic disk drive, and / or optical disk drive. The storage device 158 can include an internal storage device, an attached storage device (including detachable and non-detachable storage devices), and / or a network accessible storage device, as non-limiting examples.

[0098] System 150 includes an encoder / decoder module 156 configured, for example, to process data to provide an encoded video or decoded video, and the encoder / decoder module 156 can include its own processor and memory. The encoder / decoder module 156 represents module(s) that can be included in a device to perform the encoding and / or decoding functions. As is known, a device can include one or both of the encoding and decoding modules. Additionally, encoder / decoder module 156 can be implemented as a separate element of system 150 or can be incorporated within processor 152 as a combination of hardware and software as known to those skilled in the art.

[0099] Program code to be loaded onto processor 152 or encoder / decoder 156 to perform the various aspects described in this document can be stored in storage device 158 and subsequently loaded onto memory 154 for execution by processor 152. In accordance with various embodiments, one or more of processor 152, memory 154, storage device 158, and encoder / decoder module 156 can store one or more of various items during the performance of the processes described in this document. Such stored items can include, but are not limited to, the input video, the decoded video or portions of the decoded video, the bitstream, matrices, variables, and intermediate or final results from the processing of equations, formulas, operations, and operational logic.

[0100] In some embodiments, memory inside of the processor 152 and / or the encoder / decoder module 156 is used to store instructions and to provide working memory for processing that is needed during encoding or decoding. In other embodiments, however, a memory external to the processing device (for example, the processing device can be either the processor 152 or the encoder / decoder module 152) is usedfor one or more of these functions. The external memory can be the memory 154 and / or the storage device 158, for example, a dynamic volatile memory and / or a non-volatile flash memory. In several embodiments, an external non-volatile flash memory is used to store the operating system of, for example, a television. In at least one embodiment, a fast external dynamic volatile memory such as a RAM is used as working memory for video coding and decoding operations, such as for MPEG-2 (MPEG refers to the Moving Picture Experts Group, MPEG-2 is also referred to as ISO / IEC 13818, and 13818-1 is also known as H.222, and 13818-2 is also known as H.262), HEVC (HEVC refers to High Efficiency Video Coding, also known as H.265 and MPEG-H Part 2), or VVC (Versatile Video Coding, a new standard being developed by JVET, the Joint Video Experts Team).

[0101] The input to the elements of system 150 can be provided through various input devices as indicated in block 172. Such input devices include, but are not limited to, (i) a radio frequency (RF) portion that receives an RF signal transmitted, for example, over the air by a broadcaster, (ii) a Component (COMP) input terminal (or a set of COMP input terminals), (iii) a Universal Serial Bus (USB) input terminal, and / or (iv) a High Definition Multimedia Interface (HDMI) input terminal. Other examples, not shown in FIG. 1 C, include composite video.

[0102] In various embodiments, the input devices of block 172 have associated respective input processing elements as known in the art. For example, the RF portion can be associated with elements suitable for (i) selecting a desired frequency (also referred to as selecting a signal, or band-limiting a signal to a band of frequencies), (ii) downconverting the selected signal, (iii) band-limiting again to a narrower band of frequencies to select (for example) a signal frequency band which can be referred to as a channel in certain embodiments, (iv) demodulating the downconverted and band-limited signal, (v) performing error correction, and (vi) demultiplexing to select the desired stream of data packets. The RF portion of various embodiments includes one or more elements to perform these functions, for example, frequency selectors, signal selectors, band-limiters, channel selectors, filters, downconverters, demodulators, error correctors, and demultiplexers. The RF portion can include a tuner that performs various of these functions, including, for example, downconverting the received signal to a lower frequency (for example, an intermediate frequency or a near-baseband frequency) or to baseband. In one set-top box embodiment, the RF portion and its associated input processing element receives an RF signal transmitted over a wired (for example, cable) medium, and performs frequency selection by filtering, downconverting, and filtering again to a desired frequency band. Various embodiments rearrange the order of the above-described (and other) elements, remove some of these elements, and / or add other elements performing similar or different functions. Addingelements can include inserting elements in between existing elements, such as, for example, inserting amplifiers and an analog-to-digital converter. In various embodiments, the RF portion includes an antenna.

[0103] Additionally, the USB and / or HDMI terminals can include respective interface processors for connecting system 150 to other electronic devices across USB and / or HDMI connections. It is to be understood that various aspects of input processing, for example, Reed-Solomon error correction, can be implemented, for example, within a separate input processing IC or within processor 152 as necessary. Similarly, aspects of USB or HDMI interface processing can be implemented within separate interface ICs or within processor 152 as necessary. The demodulated, error corrected, and demultiplexed stream is provided to various processing elements, including, for example, processor 152, and encoder / decoder 156 operating in combination with the memory and storage elements to process the datastream as necessary for presentation on an output device.

[0104] Various elements of system 150 can be provided within an integrated housing, Within the integrated housing, the various elements can be interconnected and transmit data therebetween using suitable connection arrangement 174, for example, an internal bus as known in the art, including the Inter- IC (I2C) bus, wiring, and printed circuit boards.

[0105] The system 150 includes communication interface 160 that enables communication with other devices via communication channel 162. The communication interface 160 can include, but is not limited to, a transceiver configured to transmit and to receive data over communication channel 162. The communication interface 160 can include, but is not limited to, a modem or network card and the communication channel 162 can be implemented, for example, within a wired and / or a wireless medium.

[0106] Data is streamed, or otherwise provided, to the system 150, in various embodiments, using a wireless network such as a Wi-Fi network, for example IEEE 802.11 (IEEE refers to the Institute of Electrical and Electronics Engineers). The Wi-Fi signal of these embodiments is received over the communications channel 162 and the communications interface 160 which are adapted for Wi-Fi communications. The communications channel 162 of these embodiments is typically connected to an access point or router that provides access to external networks including the Internet for allowing streaming applications and other over-the-top communications. Other embodiments provide streamed data to the system 150 using a set-top box that delivers the data over the HDMI connection of the input block 172. Still other embodiments provide streamed data to the system 150 using the RF connection of the input block 172. As indicated above, various embodiments provide data in a non-streaming manner. Additionally, various embodiments use wireless networks other than Wi-Fi, for example a cellular network or a Bluetooth network.

[0107] The system 150 can provide an output signal to various output devices, including a display 176, speakers 178, and other peripheral devices 180. The display 176 of various embodiments includes one or more of, for example, a touchscreen display, an organic light-emitting diode (OLED) display, a curved display, and / or a foldable display. The display 176 can be for a television, a tablet, a laptop, a cell phone (mobile phone), or other device. The display 176 can also be integrated with other components (for example, as in a smart phone), or separate (for example, an external monitor for a laptop). The other peripheral devices 180 include, in various examples of embodiments, one or more of a stand-alone digital video disc (or digital versatile disc) (DVR, for both terms), a disk player, a stereo system, and / or a lighting system. Various embodiments use one or more peripheral devices 180 that provide a function based on the output of the system 150. For example, a disk player performs the function of playing the output of the system 150.

[0108] In various embodiments, control signals are communicated between the system 150 and the display 176, speakers 178, or other peripheral devices 180 using signaling such as AV.Link, Consumer Electronics Control (CEC), or other communications protocols that enable device-to-device control with or without user intervention. The output devices can be communicatively coupled to system 150 via dedicated connections through respective interfaces 164, 166, and 168. Alternatively, the output devices can be connected to system 150 using the communications channel 162 via the communications interface 160. The display 176 and speakers 178 can be integrated in a single unit with the other components of system 150 in an electronic device such as, for example, a television. In various embodiments, the display interface 164 includes a display driver, such as, for example, a timing controller (T Con) chip.

[0109] The display 176 and speaker 178 can alternatively be separate from one or more of the other components, for example, if the RF portion of input 172 is part of a separate set-top box. In various embodiments in which the display 176 and speakers 178 are external components, the output signal can be provided via dedicated output connections, including, for example, HDMI ports, USB ports, or COMP outputs.

[0110] The system 150 may include one or more sensor devices 168. Examples of sensor devices that may be used include one or more GPS sensors, gyroscopic sensors, accelerometers, light sensors, cameras, depth cameras, microphones, and / or magnetometers. Such sensors may be used to determine information such as user's position and orientation. Where the system 150 is used as the control module for an extended reality display (such as control modules 124, 132), the user's position and orientation may be used in determining how to render image data such that the user perceives the correct portion of a virtual object or virtual scene from the correct point of view. In the case of head-mounted display devices, the position and orientation of the device itself may be used to determine the position and orientation of the user for the purpose of rendering virtual content. In the case of other display devices, such as a phone, a tablet, acomputer monitor, or a television, other inputs may be used to determine the position and orientation of the user for the purpose of rendering content. For example, a user may select and / or adjust a desired viewpoint and / or viewing direction with the use of a touch screen, keypad or keyboard, trackball joystick, or other input. Where the display device has sensors such as accelerometers and / or gyroscopes, the viewpoint and orientation used for the purpose of rendering content may be selected and / or adjusted based on motion of the display device.

[0111] The embodiments can be carried out by computer software implemented by the processor 152 or by hardware, or by a combination of hardware and software. As a non-limiting example, the embodiments can be implemented by one or more integrated circuits. The memory 154 can be of any type appropriate to the technical environment and can be implemented using any appropriate data storage technology, such as optical memory devices, magnetic memory devices, semiconductor-based memory devices, fixed memory, and removable memory, as non-limiting examples. The processor 152 can be of any type appropriate to the technical environment, and can encompass one or more of microprocessors, general purpose computers, special purpose computers, and processors based on a multi-core architecture, as non-limiting examples.

[0112] The use of 3D applications is becoming more popular every day. To be able to exploit such applications, different data formats are being used. One of the data formats is point clouds. Point clouds are a set of unordered points with coordinates (x, y, z) corresponding to a point location in space and its attributes (such as colors and normal vectors).

[0113] Use of these new types of data likely requires new compression methods to efficiently store and transmit data, especially since point clouds may have millions of points. Of the different methods that have been proposed, learning-based architectures are gaining strength. Those architectures are often extensions of learning-based methods explored in the 2D image domain.

[0114] The conference paper, L. Dihn, J. Sohl-Dickstein, and S. Bengio, Density Estimation Using Real NVP, ARXIV: 1605.08803v3 (2016), is understood to explore the use of normalizing flows as a compression architecture in the 2D image compression domain. However, this method uses a squeeze layer that is not adapted for point cloud data structure. Normalizing flows are a type of architecture that produces a latent space from an input. The latent space is a representation of the input with different coefficients. The goal is to produce a latent space that is easier to compress than the original input.

[0115] Two architectures modified example embodiments in accordance with the ‘317 application to perform compression are presented herein. For some embodiments, the goal of these new architectures is to enhance the performance and reduce the complexity of the network, such as memory complexity. Someexample implementations in accordance with example embodiments of the ‘317 application have 278 million parameters.

[0116] FIG. 2A is a block diagram showing an example system according to some embodiments. FIG. 2A illustrates a block diagram of a system within which aspects of the present embodiments may be implemented, according to another embodiment. FIG. 2A shows one embodiment of an apparatus 200 for encoding or decoding a point cloud or attributes of a point cloud as described according to any one of the embodiments described herein. The apparatus may include a processor 210 and can be interconnected to a memory 220 through at least one port. Both processor 210 and memory 220 can also have one or more additional interconnections to external connections.

[0117] Processor 220 is also configured to code one or more attributes of a point cloud using an invertible neural network, using any one of the embodiments described herein. For instance, the processor 210 is configured using a computer program product comprising code instructions that implements any one of embodiments described herein.

[0118] FIG. 2B is a system diagram illustrating an example set of interfaces for networking according to some embodiments. In an embodiment, illustrated in FIG. 2B, in a transmission context between two remote devices A and B 230, 234 over a communication network NET 232, the device A 230 comprises a processor in relation with memory RAM and ROM which are configured to implement a method for encoding a point cloud, as described with FIGs. 3-13 and the device B comprises a processor in relation with memory RAM and ROM which are configured to implement a method for decoding a point cloud as described in relation with FIGs 3-13. In accordance with an example, the network is a broadcast network, adapted to broadcast / transmit encoded point cloud from device A to decoding devices including the device B.

[0119] FIG. 2C is a message bitfield diagram showing an example packet according to some embodiments. FIG. 2C shows an example of the syntax of a signal transmitted over a packet-based transmission protocol. Each transmitted packet P comprises a header H 260 and a payload PAYLOAD 262. In some embodiments, the payload PAYLOAD 262 may comprise coded point cloud data according to any one of the embodiments described above. In a variant, the signal comprises a flag indicating a deep learning method for decoding the point cloud or for decoding one or more attributes of the point cloud.

[0120] FIG. 3 is a process diagram showing an example normalizing flow (NF) architecture to compress and uncompress point cloud data according to some embodiments. The inversibility of a normalizing flow architecture enables a normalizing flow architecture to differ from a lot of other architecture types. After a latent space is created, the original input may be reconstructed by applying the architecture in an inverted or reverse fashion. This property is very interesting considering that data compression may be naturally treatedas an inversion problem. To be able to use such architectures, a squeezing operation is performed, efficiently trading the spatial size of the input for channels in order for the operations to be performed. An invertible neural network (flow) block was used in the ‘317 application and the journal article, Pinheiro, R. Borba, et al., NF-PCAC: Normalizing Flow Based Point Cloud Attribute Compression, 2023 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP) (2023) ‘Pinheiro"). The architecture was adapted from the article Xie, Y., Cheng, K. L, and Chen, Q., Enhanced Invertible Encoding for Learned Image Compression, ARXIV:2108.03690V1 (2021) {‘Xie"). The architecture is shown in FIG. 3.

[0121] FIG. 3 illustrates encoding and decoding processes with a bitstream generated in-between each process. FIG 3 shows a normalizing flow architecture adapted for compression of 3D point cloud color attributes. For some embodiments, a point cloud is inputted into an example architecture 300, which includes a feature enhancement block 302, an invertible neural network (INN) (or flow) block 304, a channel average block 306, and an attention layer 308.

[0122] On the encoding side, the normalizing flow architecture produces a latent space that is encoded by an entropy encoder 312 to produce a bitstream. In the example of FIG. 3, the entropy encoder 312 is a neural network-based encoder coupled with a hyperprior encoder 310. For the example configuration shown in FIG. 3, the input to entropy encoder 312 receives the output of the attention layer 308 and the previous version of the output of the hyperprior decoder 314. The hyperprior encoder 310 may include a neural network with sparse convolutions to transform the output of the attention layer 308 into a bitstream with side information. This side information is used by the entropy encoder 312 to output the main bitstream. For some embodiments, the goal of the hyperprior encoder 310 is to generate side information for use by the (main) entropy encoder. For some embodiments, the output of the hyperprior encoder 310 is passed through a hyperprior decoder 314, the output of which is used as an input into the entropy encoder 312. The output of the attention layer 308 corresponds to the latent space of the original signal. The output of the hyperprior decoder 314, after passing through the hyperprior encoder 310, will provide context information to code the original latent space. In terms of data shape, for each coefficient in the latent space the hyperprior decoder output may provide a mean and a scale. Hence, the hyperprior may have double the size of the latent space that is outputted by the attention layer 308.

[0123] On the decoder side, the bitstream is entropy-decoded using, for instance, a neural network-based entropy decoder 316 coupled with a hyperprior decoder 314 to provide a decoded latent space. The decoded latent is passed to the example normalizing flow architecture that includes an attention layer 318, a channel copy layer 320, an invertible neural network 304, and a feature enhancement block 322 to produce a reconstructed point cloud.

[0124] For some embodiments, the feature enhancement layer has the goal of extracting more non-linear features from the original point cloud. In the architecture shown in FIG. 3, the feature enhancement layer provides features to the INN block 304.

[0125] For some embodiments, the invertible neural network 304 includes three sets of a voxel shuffling layer, a 1x1 convolution, and a set of coupling layers. Because each of the layers has its own weight and bias, the three sets are not represented as a loop repeated three times. Since the data goes through a voxel shuffling layer in each set, the number of channels changes. As explained below, the size of the filters in the subsequent convolutions also changes. In the architecture ofX / e, there is a sequence of 4 repetitions of the invertible block, which comprises a pixel shuffling layer, a 1x1 convolution, and 3 coupling layers.

[0126] In the architecture of FIG. 3, the number of invertible blocks is reduced in comparison with other architectures, such as example architectures shown in the ‘317 application. This reduction is motivated by the evolution from a 2D architecture to a 3D domain. When adding a new dimension without reducing the number of repetitions of the invertible block, the number of coefficients explodes, and the use of the network may be affected negatively. The number of coupling layers in the invertible block also may be reduced for the same reason. For example, one squeeze operation may include one 1x1 sparse convolution followed by two coupling layers.

[0127] For the INN to have the desired effect and the convergence to be sped up, the voxel shuffling layer may be specifically designed for sparse 3D data. For some embodiments, the voxel shuffling layer has the goal of efficiently trading spatial dimension for channels without losing any information. For some embodiments, the 1x1 convolution of the invertible block has the goal of enhancing feature representation for the coupling layers. In the INN illustrated on FIG. 3, the 1x1 convolution is a sparse 1x1 convolution instead of the 1x1 convolution shown in Xie. A coupling layer may be a series of inversible transformations that are applied to an input tensor. In FIG. 3, for some embodiments, all the convolutions used in the coupling layers may be sparse 3D convolutions.

[0128] For some embodiments, the channel average layer is a layer in which the number of channels for the latent space is reduced by taking the average of all the channels in each spatial location. The voxel shuffling layer may be specifically designed for 3D sparse data. Without such a design for 3D sparse data, the channel average may take into account several zeros, thereby passing distorted coefficients to the attention layer.

[0129] For some embodiments, the attention layer has the goal of focusing on "more important” areas in the point cloud. For some embodiments, the attention layer uses a sigmoid function as part of a weighting function for the encoder to allocate more bits to certain regions of the point cloud data. The attention layerblock illustrated in FIG. 3 is different from Xie because all the regular 2D convolutions of Xie are replaced with sparse 3D convolutions, which enables the process to handle the sparsity and the extra dimension of the point clouds.

[0130] The sparse nature and high number of points of a point cloud representation are typically not adapted to the use of regular 3D convolutions. Therefore, the architecture illustrated in FIG. 3 uses sparse convolutions. In some embodiments, a specifically designed 3D voxel shuffling layer is provided, which allows the system illustrated in FIG. 3 to converge faster.

[0131] For some embodiments, the layers of a network, which may become very large with over 270 million parameters, may not necessarily be used in compression. For some embodiments, a goal may be to enhance the performance of the architecture in the high bitrate domain while also reducing the number of coefficients.

[0132] In this regard, two new architectures, which are adapted from examples of normalizing flow based point cloud attribute compression, e.g., as shown in Pinheiro and, e.g., in accordance with example embodiments of the ‘317 application, reduce the size of the network. Such architectures may be friendly for standardization purposes without losing performance. These architectures may improve performance compared to other models and offer better control over the number of channels in the model. Such control is thought not to be possible with previous normalizing flow architectures.

[0133] The first new architecture, while not adding any new blocks compared to, e.g., the architecture shown in FIG. 3, does increase the size of the hyperprior model and exclude the averaging layer and the last block in the core of a normalizing flow (NF) architecture.

[0134] The second architecture uses a strategy used in the super resolution field called back projection according to Haris, M., Shakhnarovich, G., and Ukita, N., Deep Back-Projection Networks for SuperResolution, 2018 IEEE / CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2018) tj'Haris"). This second architecture enables better control of the number of channels, while trying to avoid loss of information. Both the first and second architectures are adapted to code point cloud attributes.

[0135] FIG. 4A is a process diagram illustrating an example feature enhancement layer according to some embodiments. The feature enhancement layer 400 has the goal to help extract more non-linear features from the original point cloud. An example of a feature enhancement layer is illustrated in FIG. 4A, which is composed of a dense block architecture 402 followed by three 3D Sparse Convolutions of kernel size 3x3x3 (404, 406, 408) followed by another dense block architecture 410.

[0136] FIG. 4B is a process diagram illustrating an example feature enhancement according to some embodiments. An example of a sparse dense block 400 that may be used in the feature enhancement layer is illustrated in FIG. 4B. The dense block 400 has the goal of preserving initial features along the convolutions by concatenating (cat) the output of previous convolutions in the output of current convolution. This block has been proven to enhance performance of learning-based architecture. In the architecture illustrated in FIG. 7, the feature enhancement layer provides better features for the INN in the core of the illustrated network. The feature enhancement layer is inspired by the one from the architecture of Xie, but where sparse 3D convolutions are used instead of the 2D regular convolutions of Xie. For some embodiments, a convolution may be done by a 3D Sparse Convolution 3x3x3 block 422, 426, 430, 434, 438. For some embodiments, a LeakyReLU block 424, 428, 432, 436 may be used between a 3D Sparse Convolution 3x3x3 block 422, 426, 430, 434, 438 and the concatenating (cat) blocks.

[0137] FIG. 4C is a process diagram illustrating an example coupling layer according to some embodiments. A coupling layer 450 is a series of transformations that are applied to an input tensor and are completely inversible. The input tensor 452 is split into 2 parts and each part goes through its own transformation according to a scheme illustrated on FIG. 4C. On FIG. 4C, the input tensor 452 is split in 2 parts: xi (456) and X2 (454, 458), which are then transformed in yi (460, 464) and y2 (462) before being concatenated together again to provide an output y 470. The transformations Gi (476), G2 (472), Hi (478), and H2 (474) are all composed of 3 sparse 3D convolutions each.

[0138] FIG. 4D is a process diagram illustrating an example transformation block for coupling layers according to some embodiments. FIG. 4D illustrates an example of a transformation block 480 for the coupling layers that may be used for the transformations G1, G2, Hi and H2. Inside a coupling layer, each one of the transformations is composed of 3 stages of 3D sparse convolutions of kernel 3x3x3 (482, 486, 490) and a LeakyReLU 484, 488. The LeakyReLU is a deep learning block. The LeakyReLU block is similar to a ReLU block except that instead of passing only positive values, a LeakyReLU also "leaks” negative values as well. For some embodiments, the amount of "leak” may be indicated with a parameter.

[0139] The arrangement of transformations guarantees the invertibility. In the embodiment illustrated on FIG. 2 and 4C, all the convolutions used in the coupling layers are sparse 3D convolutions.

[0140] The channel average layer is a layer where the number of channels for the latent space is reduced by taking the average of all the channels in each spatial location. The voxel shuffling layer specifically designed for 3D sparse data (which is described further below) is especially important here, because without it, the channel average would have several zeros taken into account, passing distorted coefficients to the attention layer.

[0141] The attention layer has the goal of helping the architecture to focus on more important areas in the point cloud. It uses a sigmoid function to give a weight to tell the encoder which regions of the point cloud would need more bits to be encoded. The attention layer block illustrated on FIG. 2 is also modified by replacing all the regular 2D convolutions of Xie by sparse 3D convolutions to be able to handle the sparsity and the extra dimension of the point clouds.

[0142] The sparse nature and high number of points of the point cloud representation are not adapted to the use of regular 3D convolutions. Therefore, the architecture illustrated on FIG. 2 uses sparse convolutions.

[0143] FIG. 5 is a flowchart illustrating an example process for encoding a point cloud according to some embodiments. An embodiment for coding one or more attributes of the point cloud using an invertible neural network is illustrated on FIG. 5. FIG. 5 illustrates an example of a block diagram of a method for encoding one or more attributes of the point cloud. At 502, the point cloud is provided as input to the encoding system. The encoding system comprises at least an invertible neural network configured for encoding at least one attribute of the point cloud. In some variants, the encoding system comprises geometry encoding module configured for encoding geometry of the point cloud. At 504, a latent representation of at least one attribute of the point cloud is obtained using at least the invertible neural network. At 506, the latent is encoded to produce a bitstream, for instance using a neural network-based entropy encoder.

[0144] FIG. 6 is a flowchart illustrating an example process for reconstructing a point cloud according to some embodiments. Another embodiment for coding one or more attributes of the point cloud using an invertible neural network is illustrated on FIG. 6. FIG. 6 illustrates an example of a block diagram of a method for decoding one or more attributes of the point cloud. At 602, a bitstream is provided to the decoding system. The bitstream comprises at least coded data representative of at least one attribute of the point cloud. The decoding system comprises at least an invertible neural network configured for decoding at least one attribute of the point cloud. In some variants, the bitstream also comprises coded data representative of the geometry of the point cloud and the decoding system comprises geometry decoding module configured for decoding and reconstructing the geometry of the point cloud. At 604, a latent representation of at least one attribute of the point cloud is obtained by decoding the bitstream part representative of the at least one attribute, for instance using a neural network-based entropy decoder. At 606, the at least one attribute is reconstructed using at least the invertible neural network.Normalizing Flow Non-Averaged (NF-No-Avg) Architecture

[0145] FIG. 7 is a process diagram showing an example normalizing flow, non-averaged process according to some embodiments. In some embodiments, such example processes may be referred to as"NF-No Avg” for ease of description, although a variety of designs and architectures are contemplated. The architecture 700 is a smaller version of the architecture shown in FIG. 3. For some embodiments, the main component in the initial architecture of FIG. 3 that caused non-invertibility was the channel average layer. In the example of FIG. 3, the channel averaging block is connected to the output of the INN block, which may be designated as the NF core for some embodiments. This strong non-invertibility aspect of the network may cause a saturation of the results for high bitrates. This saturation refers to the rate-distortion curve. Negatively, even if a very large bitrate is allocated for a given point-cloud, the reconstruction quality will not change that much and may have diminished rate distortion performance. Combining the channel averaging layer and a replacement of empty voxels with the average of their neighbors results in a sort of smoothing of the attributes of the original point cloud. The replacement of empty voxels may happen in a sparse voxel shuffling block. A sparse voxel shuffling block may be used in various architectures and may contribute to the performance. In the example of FIG. 3, sparse voxel shuffling is done in the voxel shuffling layer.

[0146] The voxel shuffling layer shown in FIG. 3 is kept intact in this new architecture of FIG. 7. However, the last group of blocks of the INN 704, which includes a voxel shuffling layer, a 1x1 sparse convolution, and a set of coupling layers, is removed to limit an increase in the number of channels. Doing so enables removal of the channel average layer and maintains a reasonable amount of computational complexity and number of coefficients. The example architecture of FIG. 7 according to some implementations has 25 million parameters, which is less than 10% of the 278 million parameters of FIG. 3 according to some implementations.

[0147] For some embodiments, a point cloud is inputted into an example architecture 700, which includes a feature enhancement block 702, an invertible neural network (INN) (or flow) block 704, and an attention layer 706. For the example configuration shown in FIG. 7, the input to entropy encoder 708 receives the output of the attention layer 706 and the previous version of the output of the hyperprior decoder 712. For some embodiments, the output of the hyperprior encoder 710 is passed through a hyperprior decoder 712, the output of which is used as an input into the entropy encoder 708.

[0148] On the decoder side, the bitstream is entropy-decoded using, for instance, a neural network-based entropy decoder 714 coupled with a hyperprior decoder 712 to provide a decoded latent space. The decoded latent is passed to the example normalizing flow architecture that includes an attention layer 716, an invertible neural network 704, and a feature enhancement block 718 to produce a reconstructed point cloud.Normalizing Flow Back Projection (NF-BP) Architecture

[0149] FIGs. 8A is a process diagram showing an example average back projection process according to some embodiments. FIGs. 8B is a process diagram showing an example copy back projection process according to some embodiments. A second architecture uses an average back projection block and a copy back projection block, as shown in FIGs. 8A and 8B. For some embodiments, these blocks have the goal of reducing the number of channels in the inner layers of the normalizing flow block (or core), while trying to keep all the needed information in the channels. The average back projection block and a copy back projection block are used in super resolution architectures to enhance the reconstruction with details that were lost in the averaging portion.

[0150] The "Av J,” (down arrow) block 804 of FIG. 8A performs an averaging over the channels of the tensor, reducing the dimensionality. The "Cp f (up arrow) block 806, 854 of FIG. 8A and 8B does the inverse and copies the channels to recover the same dimensionality as before.

[0151] In the architecture 800 of FIG. 8A, the output of the "Cp f block 806 is subtracted from the original input 802 to form an input into the "CsJ,” block 808. The "CsJ,” block 808 performs a down sampling, taking into consideration the residual produced by the difference between the original input tensor and the tensor that would be reconstructed in the copying after averaging. The output of the "CsJ,” block 808 is added to the averaged tensor to produce the output 810 of the ABP block. For some embodiments, the "CsJ,” block 808 may include convolutional layers that produce residuals (part of the "CsJ,” block output) to be added to the averaged tensor (“Av J,” block output) to produce the output of the average back projection block. For some embodiments, the goal is to obtain a down sampled tensor (“CsJ,” block output) that is more representative of the original data than a plain average.

[0152] The copy back projection block architecture 850 of FIG. 8B does the inverse of the average back projection block of FIG. 8A. As a result, an up sampled version of the data is generated as the output 860. The output of the "CsJ,” block 856 is subtracted from the original input 852 to form an input into the "Csf block 858. The "Csf block 858 performs an up sampling, taking into consideration the residual produced by the difference between the original input tensor and the tensor that would be reconstructed in the downsampling after copying. The output of the "Csf block 858 is added to the copy tensor to produce the output 860 of the CBP block. For some embodiments, the "Csf block 858 may include convolutional layers that produce residuals (part of the "Csf block output) to be added to the copy tensor ("Cpf block output) to produce the output of the up sampled tensor ("Csf block output) that is more representative of the original data than a plain copy.

[0153] FIG. 9 is a process diagram showing an example normalizing flow back projection process according to some embodiments. In some embodiments, such an example process 900 may be referred toas"NF-BP” for ease of description, although a variety of designs and architectures are contemplated. On the encoding side of the bitstream in FIG. 9 an average back projection (ABP) block is inserted after each sequence of voxel shuffling, 1x1 sparse convolution, and set of coupling layers. On the decoding side of the bitstream in FIG. 9, a copy back projection (CBP) block is inserted before each sequence of voxel shuffling, 1x1 sparse convolution, and set of coupling layers.

[0154] These ABP and CBP blocks help reduce the size of the original model by progressively reducing the number of channels along the core of the network, which may be for each sequence of voxel shuffling, 1x1 sparse convolution, and set of coupling layers. The example architecture of FIG. 9 according to some implementations uses 18 million coefficients, which is also less than 10% of the original 278 million parameters according to some implementations of FIG. 3. For some embodiments, only two sequences of voxel shuffling, 1x1 sparse convolution, a set of coupling layers, and an ABP or CBP block are performed (instead of the three sequences shown in FIG. 9). For some embodiments, four or more sequences of voxel shuffling, 1x1 sparse convolution, a set of coupling layers, and an ABP or CBP block are performed (instead of the three sequences shown in FIG. 9).

[0155] For some embodiments, a point cloud is inputted into an example architecture 900, which includes a feature enhancement block 902, an invertible neural network (INN) (or flow) block 904, and an attention layer 906. For the example configuration shown in FIG. 9, the input to entropy encoder 908 receives the output of the attention layer 906 and the previous version of the output of the hyperprior decoder 912. For some embodiments, the output of the hyperprior encoder 910 is passed through a hyperprior decoder 912, the output of which is used as an input into the entropy encoder 908.

[0156] On the decoder side, the bitstream is entropy-decoded using, for instance, a neural network-based entropy decoder 914 coupled with a hyperprior decoder 912 to provide a decoded latent space. The decoded latent is passed to the example normalizing flow architecture that includes an attention layer 916, an invertible neural network 904, and a feature enhancement block 918 to produce a reconstructed point cloud.Comparison with the Original Architecture

[0157] FIG. 10 is a graph 1000 illustrating an example peak signal-to-noise ratio (PSNR) vs. bits per input point for a first test scenario according to some embodiments. FIG. 11 is a graph 1100 illustrating an example peak signal-to-noise ratio (PSNR) vs. bits per input point for a second test scenario according to some embodiments. FIG. 12 is a graph 1200 illustrating an example peak signal-to-noise ratio (PSNR) vs. bits per input point for a third test scenario according to some embodiments. See d'Eon, E., et al., 8 / Voxelized Full Bodies - A Voxelized Point Cloud Dataset, ISO / IEC JTC1 / SC29 JOINT WG11 / WG1 (MPEG / JPEG) INPUTDOCUMENT WG1 1 M40059 / WG1 M74006 (2017) regarding FIGs. 10 and 12. See Xu, Y., et al., Owlii Dynamic Human Mesh Sequence Dataset, ISO / IEC JTC1 / SC29 / WG11 M41658, 120TH MPEG MEETING (2017) regarding FIG. 11.

[0158] The NF-No Avg architecture shown in FIG. 7 outperforms other learning-based processes as applied in the high bitrate domain. The removal of the averaging allows the network to perform well in the high bitrate range. However, for low bitrates, other methods may outperform an NF-No Avg architecture. This result may be due to the creation of extra, unnecessary coefficients by the voxel shuffling that were not filtered in the channel averaging. For some embodiments, the GPCC curve of FIG. 10 may be used as a base with, say, the first 2 or 3 data points in the graph being low bitrate and the last 2 or 3 data points being high bitrate. For example, in FIG. 10, the three data points of the GPCC curve that are less than 0.1 bits per input point may be considered low bitrate, and the three data points of the GPCC curve that are more than 0.35 may be considered high bit rate.

[0159] On the other hand, an NF-BP architecture performs better in the low bitrate domain but is outperformed in a high bitrate domain. The progressive averaging of the channels through the network may cause a loss of texture details of point clouds.Signaling / Syntax

[0160] For a Rate Distortion Optimization (RDO) approach, different configurations are tested for a particular bitrate. For a codec that performs an RDO approach, good performance may be achieved over a wide range of bitrates while having a network that is smaller (e.g., less parameters) than the network of Pinheiro. For some embodiments, two network configurations are tested, and the network configuration with the better (or best) reconstruction quality is chosen. An extra bit may be added to indicate to the decoder side which architecture was used on the encoder side. For some embodiments, a signaling flag (which may be one or more bits) may be used to indicate the architecture configuration. For some embodiments, the normalizing flow blocks of ‘352 application may be replaced with the configurations shown in FIGs. 7 to 9.

[0161] Current activities in MPEG AI-PCC are working on defining Al models for compression and decompression of point cloud geometry and photometry. The MPEG group currently handles photometry separately from geometry. The processes described herein add to the number of available architectures for compression to perform encoding / decoding of photometry. For example, processes described herein may be applied to the G-PCC coding standard extension based on deep learning.

[0162] The processes described herein may become an option for a codec in an RDO fashion by choosing an encoder that results in better rate-distortion. A signaling flag may be added to a bitstream to indicate whichdeep decoder architecture to use. Table 1 shows an example mapping of architectures to use for a point cloud (PC)-learning based bitstream.

[0163] An NF-No-Avg architecture is an architecture for point clod compression using sparse convolutions and a smaller version of a normalizing architecture. An NF-BP architecture is an architecture for point cloud compression using average back projection (ABP) and copy back projection (CBP) to increase reconstruction quality.

[0164] FIG. 13 is a flowchart illustrating an example process for encoding a point cloud according to some embodiments. For some embodiments, an example process 1300 may include obtaining 1302 information corresponding to a point cloud. For some embodiments, the example process 1300 may further include performing 1304 feature enhancement on point cloud data corresponding to the information. For some embodiments, the example process 1300 may further include generating 1306 data corresponding to a latent space based on the information corresponding to the point cloud. For some embodiments of the example process 1300, generating 1308 the data corresponding to the latent space may include: using the feature enhanced point cloud data as input data to a first loop through a subprocess; performing the subprocess twice, wherein the subprocess may include: passing the input data through a voxel shuffling layer to generate a voxel shuffling layer output; performing a convolution on the voxel shuffling layer output to generate convolution output data; passing the convolution output data through one or more coupling layers to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and passing the second loop subprocess output through an attention layer to generate the data corresponding to the latent space. For some embodiments, the example process 1300 may further include encoding 1310 the data corresponding to the latent space as a bitstream.

[0165] FIG. 14 is a flowchart illustrating an example process for decoding a point cloud according to some embodiments. For some embodiments, an example process 1400 may include obtaining 1402 an encoded bitstream. For some embodiments, the example process 1400 may further include decoding 1404 the encoded bitstream to generate a latent space. For some embodiments, the example process 1400 may further include reconstructing 1406 a point cloud from the generated latent space. For some embodiments of the example process 1400, reconstructing 1408 the point cloud from the generated latent space may include: passing the generated latent space through an attention layer to generate input data to a subprocess; performing the subprocess twice, wherein the subprocess may include: passing the input data through one or more coupling layers to generate a coupling layer output; and performing a convolution on the coupling layer output to generate convolution output data; passing the convolution output data through a voxel shuffling layer to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess. For some embodiments, the example process 1400 may further include performing 1410 feature enhancement on the second loop subprocess output to generate the reconstructed point cloud.

[0166] While the methods and systems in accordance with some embodiments are generally discussed in context of extended reality (XR), some embodiments may be applied to any XR contexts such as, e.g., virtual reality (VR) / mixed reality (MR) / augmented reality (AR) contexts. Also, although the term "head mounted display (HMD)” is used herein in accordance with some embodiments, some embodiments may be applied to a wearable device (which may or may not be attached to the head) capable of, e.g., XR, VR, AR, and / or MR for some embodiments.

[0167] An example method in accordance with some embodiments may include: obtaining information corresponding to a point cloud; performing feature enhancement on point cloud data corresponding to the information; generating data corresponding to a latent space based on the information corresponding to the point cloud, wherein generating the data corresponding to the latent space may include: using the feature enhanced point cloud data as input data to a first loop through a subprocess; performing the subprocess twice, wherein the subprocess may include: passing the input data through a voxel shuffling layer to generate a voxel shuffling layer output; performing a convolution on the voxel shuffling layer output to generate convolution output data; passing the convolution output data through one or more coupling layers to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and passing the second loop subprocess output through an attention layer to generate the data corresponding to the latent space; and encoding the data corresponding to the latent space as a bitstream.

[0168] For some embodiments of the example method, the subprocess may be performed three times:

[0169] For some embodiments of the example method, the subprocess further includes performing an average back projection process on an output of the one or more coupling layers.

[0170] For some embodiments of the example method, performing the average back projection process may include performing at least one of an averaging process, a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

[0171] For some embodiments of the example method, the information corresponding to the point cloud further corresponds to one or more voxels.

[0172] For some embodiments of the example method, the subprocess is a normalizing flow (NF) subprocess.

[0173] For some embodiments of the example method, the encoded bitstream may include a signaling flag to indicate an architecture configuration.

[0174] An example apparatus in accordance with some embodiments may include: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform any of the methods listed above.

[0175] A second example method in accordance with some embodiments may include: obtaining information corresponding to a point cloud; generating data corresponding to a latent space based on the information corresponding to the point cloud, wherein generating the data corresponding to the latent space may include: using the feature enhanced point cloud data as input data to a first loop through a subprocess; performing the subprocess twice, wherein the subprocess may include: passing the input data through a voxel shuffling layer to generate a voxel shuffling layer output; performing a convolution on the voxel shuffling layer output to generate convolution output data; passing the convolution output data through one or more coupling layers to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and generating the data corresponding to the latent space based on the second loop subprocess output; and encoding the data corresponding to the latent space as a bitstream.

[0176] For some embodiments of the second example method, the subprocess is performed three times:

[0177] For some embodiments of the second example method, the subprocess further includes performing an average back projection process on an output of the one or more coupling layers.

[0178] For some embodiments of the second example method, performing the average back projection process may include performing at least one of an averaging process, a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

[0179] For some embodiments of the second example method, the information corresponding to the point cloud further corresponds to one or more voxels.

[0180] For some embodiments of the second example method, the subprocess is a normalizing flow (NF) subprocess.

[0181] For some embodiments of the second example method, the encoded bitstream may include a signaling flag to indicate an architecture configuration.

[0182] A second example apparatus in accordance with some embodiments may include: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform any of the methods listed above.

[0183] A third example method in accordance with some embodiments may include: obtaining an encoded bitstream; decoding the encoded bitstream to generate a latent space; and reconstructing a point cloud from the generated latent space, wherein reconstructing the point cloud from the generated latent space may include: passing the generated latent space through an attention layer to generate input data to a subprocess; performing the subprocess twice, wherein the subprocess may include: passing the input data through one or more coupling layers to generate a coupling layer output; performing a convolution on the coupling layer output to generate convolution output data; passing the convolution output data through a voxel shuffling layer to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and performing feature enhancement on the second loop subprocess output to generate the reconstructed point cloud.

[0184] For some embodiments of the third example method, the subprocess may be performed three times:

[0185] For some embodiments of the third example method, the subprocess further may include performing a copy back projection process on an output of the voxel shuffling layer.

[0186] For some embodiments of the third example method, performing the copy back projection process may include performing at least one of a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

[0187] For some embodiments of the third example method, the information corresponding to the point cloud further corresponds to one or more voxels.

[0188] For some embodiments of the third example method, the subprocess is a normalizing flow (NF) subprocess.

[0189] For some embodiments of the third example method, the encoded bitstream may include a signaling flag to indicate an architecture configuration.

[0190] A third example apparatus in accordance with some embodiments may include: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform any of the methods listed above.

[0191] A fourth example method / apparatus in accordance with some embodiments may include: obtaining an encoded bitstream; decoding the encoded bitstream to generate a latent space; and reconstructing a point cloud from the generated latent space, wherein reconstructing the point cloud from the generated latent space may include: obtaining input data to a subprocess based on the generated latent space; performing the subprocess twice, wherein the subprocess may include: passing the input data through one or more coupling layers to generate a coupling layer output; performing a convolution on the coupling layer output to generate convolution output data; passing the convolution output data through a voxel shuffling layer to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and generating the reconstructed point cloud based on the second loop subprocess output.

[0192] For some embodiments of the fourth example method, the subprocess may be performed three times:

[0193] For some embodiments of the fourth example method, the subprocess further may include performing a copy back projection process on an output of the voxel shuffling layer.

[0194] For some embodiments of the fourth example method, performing the copy back projection process includes performing at least one of a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

[0195] For some embodiments of the fourth example method, the information corresponding to the point cloud further corresponds to one or more voxels.

[0196] For some embodiments of the fourth example method, the subprocess is a normalizing flow (NF) subprocess.

[0197] For some embodiments of the fourth example method, the encoded bitstream may include a signaling flag to indicate an architecture configuration.

[0198] A fourth example apparatus in accordance with some embodiments may include: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform any of the methods listed above.

[0199] Another example apparatus in accordance with some embodiments may include at least one processor configured to perform one of the methods listed above.

[0200] Another example apparatus in accordance with some embodiments may include a computer- readable medium storing instructions for causing one or more processors to perform one of the methods listed above.

[0201] Another example apparatus in accordance with some embodiments may include at least one processor and at least one non-transitory computer-readable medium storing instructions for causing the at least one processor to perform any one of the methods listed above.

[0202] An example computer-readable medium in accordance with some embodiments may include storing a scene description file generated according to one of the methods listed above.

[0203] An example signal in accordance with some embodiments may include a scene description file generated according to any one of the methods listed above.

[0204] This disclosure describes a variety of aspects, including tools, features, embodiments, models, approaches, etc. Many of these aspects are described with specificity and, at least to show the individual characteristics, are often described in a manner that may sound limiting. However, this is for purposes of clarity in description, and does not limit the disclosure or scope of those aspects. Indeed, all of the different aspects can be combined and interchanged to provide further aspects. Moreover, the aspects can be combined and interchanged with aspects described in earlier filings as well.

[0205] The aspects described and contemplated in this disclosure can be implemented in many different forms. While some embodiments are illustrated specifically, other embodiments are contemplated, and the discussion of particular embodiments does not limit the breadth of the implementations. At least one of the aspects generally relates to video encoding and decoding, and at least one other aspect generally relates to transmitting a bitstream generated or encoded. These and other aspects can be implemented as a method, an apparatus, a computer readable storage medium having stored thereon instructions for encoding or decoding video data according to any of the methods described, and / or a computer readable storage medium having stored thereon a bitstream generated according to any of the methods described.

[0206] In the present disclosure, the terms "reconstructed” and "decoded” may be used interchangeably, the terms "pixel” and "sample” may be used interchangeably, the terms "image,” "picture” and "frame” may be used interchangeably. Usually, but not necessarily, the term "reconstructed” is used at the encoder side while "decoded” is used at the decoder side.

[0207] The terms HDR (high dynamic range) and SDR (standard dynamic range) often convey specific values of dynamic range to those of ordinary skill in the art. However, additional embodiments are also intended in which a reference to HDR is understood to mean "higher dynamic range” and a reference to SDR is understood to mean "lower dynamic range.” Such additional embodiments are not constrained by any specific values of dynamic range that might often be associated with the terms "high dynamic range” and "standard dynamic range.”

[0208] Various methods are described herein, and each of the methods comprises one or more steps or actions for achieving the described method. Unless a specific order of steps or actions is required for proper operation of the method, the order and / or use of specific steps and / or actions may be modified or combined. Additionally, terms such as "first”, "second”, etc. may be used in various embodiments to modify an element, component, step, operation, etc., such as, for example, a "first decoding” and a "second decoding”. Use of such terms does not imply an ordering to the modified operations unless specifically required. So, in this example, the first decoding need not be performed before the second decoding, and may occur, for example, before, during, or in an overlapping time period with the second decoding.

[0209] Various numeric values may be used in the present disclosure, for example. The specific values are for example purposes and the aspects described are not limited to these specific values.

[0210] Embodiments described herein may be carried out by computer software implemented by a processor or other hardware, or by a combination of hardware and software. As a non-limiting example, the embodiments can be implemented by one or more integrated circuits. The processor can be of any type appropriate to the technical environment and can encompass one or more of microprocessors, general purpose computers, special purpose computers, and processors based on a multi-core architecture, as nonlimiting examples.

[0211] Various implementations involve decoding. "Decoding”, as used in this disclosure, can encompass all or part of the processes performed, for example, on a received encoded sequence in order to produce a final output suitable for display. In various embodiments, such processes include one or more of the processes typically performed by a decoder, for example, entropy decoding, inverse quantization, inverse transformation, and differential decoding. In various embodiments, such processes also, or alternatively, include processes performed by a decoder of various implementations described in this disclosure, for example, extracting a picture from a tiled (packed) picture, determining an upsampling filter to use and then upsampling a picture, and flipping a picture back to its intended orientation.

[0212] As further examples, in one embodiment "decoding” refers only to entropy decoding, in another embodiment "decoding” refers only to differential decoding, and in another embodiment "decoding” refers toa combination of entropy decoding and differential decoding. Whether the phrase "decoding process” is intended to refer specifically to a subset of operations or generally to the broader decoding process will be clear based on the context of the specific descriptions.

[0213] Various implementations involve encoding. In an analogous way to the above discussion about "decoding”, "encoding” as used in this disclosure can encompass all or part of the processes performed, for example, on an input video sequence in order to produce an encoded bitstream. In various embodiments, such processes include one or more of the processes typically performed by an encoder, for example, partitioning, differential encoding, transformation, quantization, and entropy encoding. In various embodiments, such processes also, or alternatively, include processes performed by an encoder of various implementations described in this disclosure.

[0214] As further examples, in one embodiment "encoding” refers only to entropy encoding, in another embodiment "encoding” refers only to differential encoding, and in another embodiment "encoding” refers to a combination of differential encoding and entropy encoding. Whether the phrase "encoding process” is intended to refer specifically to a subset of operations or generally to the broader encoding process will be clear based on the context of the specific descriptions.

[0215] When a figure is presented as a flow diagram, it should be understood that it also provides a block diagram of a corresponding apparatus. Similarly, when a figure is presented as a block diagram, it should be understood that it also provides a flow diagram of a corresponding method / process.

[0216] Various embodiments refer to rate distortion optimization. In particular, during the encoding process, the balance or trade-off between the rate and distortion is usually considered, often given the constraints of computational complexity. The rate distortion optimization is usually formulated as minimizing a rate distortion function, which is a weighted sum of the rate and of the distortion. There are different approaches to solve the rate distortion optimization problem. For example, the approaches may be based on an extensive testing of all encoding options, including all considered modes or coding parameters values, with a complete evaluation of their coding cost and related distortion of the reconstructed signal after coding and decoding. Faster approaches may also be used, to save encoding complexity, in particular with computation of an approximated distortion based on the prediction or the prediction residual signal, not the reconstructed one. A mix of these two approaches can also be used, such as by using an approximated distortion for only some of the possible encoding options, and a complete distortion for other encoding options. Other approaches only evaluate a subset of the possible encoding options. More generally, many approaches employ any of a variety of techniques to perform the optimization, but the optimization is not necessarily a complete evaluation of both the coding cost and related distortion.

[0217] The implementations and aspects described herein can be implemented in, for example, a method or a process, an apparatus, a software program, 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), the implementation of features discussed can also be implemented in other forms (for example, an apparatus or program). An apparatus can be implemented in, for example, appropriate hardware, software, and firmware. The methods can be implemented in, 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.

[0218] Reference to "one embodiment” or "an embodiment” or "one implementation” or "an implementation”, as well as other variations thereof, means that a particular feature, structure, characteristic, and so forth described in connection with the embodiment is included in at least one embodiment. Thus, the appearances of the phrase "in one embodiment” or "in an embodiment” or "in one implementation” or "in an implementation”, as well any other variations, appearing in various places throughout this disclosure are not necessarily all referring to the same embodiment.

[0219] Additionally, this disclosure may refer to "determining” various pieces of information. Determining the information can include one or more of, for example, estimating the information, calculating the information, predicting the information, or retrieving the information from memory.

[0220] Further, this disclosure may refer to "accessing” various pieces of information. Accessing the information can include one or more of, for example, receiving the information, retrieving the information (for example, from memory), storing the information, moving the information, copying the information, calculating the information, determining the information, predicting the information, or estimating the information.

[0221] Additionally, this disclosure may refer to "receiving” various pieces of information. Receiving is, as with "accessing”, intended to be a broad term. Receiving the information can include one or more of, for example, accessing the information, or retrieving the information (for example, from memory). Further, "receiving” is typically involved, in one way or another, during operations such as, for example, storing the information, processing the information, transmitting the information, moving the information, copying the information, erasing the information, calculating the information, determining the information, predicting the information, or estimating the information.

[0222] It is to be appreciated that the use of any of the following "and / or”, and "at least one of, for example, in the cases of “A / B”, "A and / or B” and "at least one of A and B”, is intended to encompass the selection of the first listed option (A) only, or the selection of the second listed option (B) only, or the selectionof both options (A and B). As a further example, in the cases of "A, B, and / or C” and "at least one of A, B, and C”, such phrasing is intended to encompass the selection of the first listed option (A) only, or the selection of the second listed option (B) only, or the selection of the third listed option (C) only, or the selection of the first and the second listed options (A and B) only, or the selection of the first and third listed options (A and C) only, or the selection of the second and third listed options (B and C) only, or the selection of all three options (A and B and C). This may be extended for as many items as are listed.

[0223] Also, as used herein, the word "signal” refers to, among other things, indicating something to a corresponding decoder. For example, in certain embodiments the encoder signals a particular one of a plurality of parameters for region-based filter parameter selection for de-artifact filtering. In this way, in an embodiment the same parameter is used at both the encoder side and the decoder side. Thus, for example, an encoder can transmit (explicit signaling) a particular parameter to the decoder so that the decoder can use the same particular parameter. Conversely, if the decoder already has the particular parameter as well as others, then signaling can be used without transmitting (implicit signaling) to simply allow the decoder to know and select the particular parameter. By avoiding transmission of any actual functions, a bit savings is realized in various embodiments. It is to be appreciated that signaling can be accomplished in a variety of ways. For example, one or more syntax elements, flags, and so forth are used to signal information to a corresponding decoder in various embodiments. While the preceding relates to the verb form of the word "signal”, the word "signal” can also be used herein as a noun.

[0224] Implementations can produce a variety of signals formatted to carry information that can be, for example, stored or transmitted. The information can include, for example, instructions for performing a method, or data produced by one of the described implementations. For example, a signal can be formatted to carry the bitstream of a described embodiment. Such a signal can be formatted, for example, as an electromagnetic wave (for example, using a radio frequency portion of spectrum) or as a baseband signal. The formatting can include, for example, encoding a data stream and modulating a carrier with the encoded data stream. The information that the signal carries can be, for example, analog or digital information. The signal can be transmitted over a variety of different wired or wireless links, as is known. The signal can be stored on a processor-readable medium.

[0225] We describe a number of embodiments. Features of these embodiments can be provided alone or in any combination, across various claim categories and types. Further, embodiments can include one or more of the following features, devices, or aspects, alone or in any combination, across various claim categories and types:• A bitstream or signal that includes one or more of the described syntax elements, or variations thereof.• A bitstream or signal that includes syntax conveying information generated according to any of the embodiments described.• Creating and / or transmitting and / or receiving and / or decoding a bitstream or signal that includes one or more of the described syntax elements, or variations thereof.• Creating and / or transmitting and / or receiving and / or decoding according to any of the embodiments described.• A method, process, apparatus, medium storing instructions, medium storing data, or signal according to any of the embodiments described.• A TV, set-top box, cell phone, tablet, or other electronic device that performs adaptation of filter parameters according to any of the embodiments described.• A TV, set-top box, cell phone, tablet, or other electronic device that performs adaptation of filter parameters according to any of the embodiments described, and that displays (e.g. using a monitor, screen, or other type of display) a resulting image.• A TV, set-top box, cell phone, tablet, or other electronic device that selects (e.g. using a tuner) a channel to receive a signal including an encoded image, and performs adaptation of filter parameters according to any of the embodiments described.• A TV, set-top box, cell phone, tablet, or other electronic device that receives (e.g. using an antenna) a signal over the air that includes an encoded image, and performs adaptation of filter parameters according to any of the embodiments described.

[0226] Note that various hardware elements of one or more of the described embodiments are referred to as "modules” that carry out (i.e., perform, execute, and the like) various functions that are described herein in connection with the respective modules. As used herein, a module includes hardware (e.g., one or more processors, one or more microprocessors, one or more microcontrollers, one or more microchips, one or more application-specific integrated circuits (ASICs), one or more field programmable gate arrays (FPGAs), one or more memory devices) deemed suitable by those of skill in the relevant art for a given implementation. Each described module may also include instructions executable for carrying out the one or more functions described as being carried out by the respective module, and it is noted that those instructions could take the form of or include hardware (i.e., hardwired) instructions, firmware instructions, software instructions, and / or the like, and may be stored in any suitable non-transitory computer-readable medium or media, such as commonly referred to as RAM, ROM, etc.

[0227] Although features and elements are described above in particular combinations, one of ordinary skill in the art will appreciate that each feature or element can be used alone or in any combination with the other features and elements. In addition, the methods described herein may be implemented in a computer program, software, or firmware incorporated in a computer-readable medium for execution by a computer or processor. Examples of computer-readable storage media include, but are not limited to, a read only memory (ROM), a random access memory (RAM), a register, cache memory, semiconductor memory devices, magnetic media such as internal hard disks and removable disks, magneto-optical media, and optical media such as CD-ROM disks, and digital versatile disks (DVDs). A processor in association with software may be used to implement a radio frequency transceiver for use in a WTRU, UE, terminal, base station, RNC, or any host computer.

Claims

CLAIMS1. A method comprising: obtaining information corresponding to a point cloud; generating data corresponding to a latent space based on the information corresponding to the point cloud, wherein generating the data corresponding to the latent space comprises: using point cloud data corresponding to the information corresponding to the point cloud as input data to a first loop through a subprocess; performing the subprocess twice, wherein the subprocess comprises: passing the input data through a voxel shuffling layer to generate a voxel shuffling layer output; performing a convolution on the voxel shuffling layer output to generate convolution output data; passing the convolution output data through one or more coupling layers to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and generating the data corresponding to the latent space based on the second loop subprocess output; and encoding the data corresponding to the latent space as a bitstream.

2. The method of claim 1 , wherein the subprocess is performed three times:

3. The method of any one of claims 1-2, wherein the subprocess further comprises performing an average back projection process on an output of the one or more coupling layers.

4. The method of any one of claims 1-3, wherein performing the average back projection process comprises performing at least one of an averaging process, a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

5. The method of any one of claims 1-4, wherein the information corresponding to the point cloud further corresponds to one or more voxels.

6. The method of any one of claims 1-5, wherein the subprocess is a normalizing flow (NF) subprocess.

7. The method of any one of claims 1 -6, wherein the encoded bitstream comprises a signaling flag to indicate an architecture configuration.

8. An apparatus comprising: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform the method of any one of claims 1 through 7.

9. A method comprising: obtaining an encoded bitstream; decoding the encoded bitstream to generate a latent space; and reconstructing a point cloud from the generated latent space, wherein reconstructing the point cloud from the generated latent space comprises: obtaining input data to a subprocess based on the generated latent space; performing the subprocess twice, wherein the subprocess comprises: passing the input data through one or more coupling layers to generate a coupling layer output; performing a convolution on the coupling layer output to generate convolution output data; passing the convolution output data through a voxel shuffling layer to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and generating the reconstructed point cloud based on the second loop subprocess output.

10. The method of claim 9, wherein the subprocess is performed three times:11 . The method of any one of claims 9-10, wherein the subprocess further comprises performing a copy back projection process on an output of the voxel shuffling layer.

12. The method of claim 11 , wherein performing the copy back projection process comprises performing at least one of a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

13. The method of any one of claims 9-12, wherein the information corresponding to the point cloud further corresponds to one or more voxels.

14. The method of any one of claims 9-13, wherein the subprocess is a normalizing flow (NF) subprocess.

15. The method of any one of claims 9-14, wherein the encoded bitstream comprises a signaling flag to indicate an architecture configuration.

16. An apparatus comprising: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform the method of any one of claims 9 through 15.

17. A method comprising: obtaining information corresponding to a point cloud; performing feature enhancement on point cloud data corresponding to the information; generating data corresponding to a latent space based on the information corresponding to the point cloud, wherein generating the data corresponding to the latent space comprises: using the feature enhanced point cloud data as input data to a first loop through a subprocess; performing the subprocess twice, wherein the subprocess comprises: passing the input data through a voxel shuffling layer to generate a voxel shuffling layer output; performing a convolution on the voxel shuffling layer output to generate convolution output data; passing the convolution output data through one or more coupling layers to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and passing the second loop subprocess output through an attention layer to generate the data corresponding to the latent space; and encoding the data corresponding to the latent space as a bitstream.

18. The method of claim 17, wherein the subprocess is performed three times:

19. The method of any one of claims 17-18, wherein the subprocess further comprises performing an average back projection process on an output of the one or more coupling layers.

20. The method of any one of claims 17-19, wherein performing the average back projection process comprises performing at least one of an averaging process, a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

21. The method of any one of claims 17-20, wherein the information corresponding to the point cloud further corresponds to one or more voxels.

22. The method of any one of claims 17-21 , wherein the subprocess is a normalizing flow (NF) subprocess.

23. The method of any one of claims 17-22, wherein the encoded bitstream comprises a signaling flag to indicate an architecture configuration.

24. An apparatus comprising: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform the method of any one of claims 17 through 23.

25. A method comprising: obtaining an encoded bitstream; decoding the encoded bitstream to generate a latent space; and reconstructing a point cloud from the generated latent space, wherein reconstructing the point cloud from the generated latent space comprises: passing the generated latent space through an attention layer to generate input data to a subprocess; performing the subprocess twice, wherein the subprocess comprises: passing the input data through one or more coupling layers to generate a coupling layer output; performing a convolution on the coupling layer output to generate convolution output data; passing the convolution output data through a voxel shuffling layer to generate a subprocess output; and using the subprocess output as the input data for a second loop through the subprocess; and performing feature enhancement on the second loop subprocess output to generate the reconstructed point cloud.

26. The method of claim 25, wherein the subprocess is performed three times:

27. The method of any one of claims 25-26, wherein the subprocess further comprises performing a copy back projection process on an output of the voxel shuffling layer.

28. The method of claim 27, wherein performing the copy back projection process comprises performing at least one of a copying process, a downsampling process, and a replacement of at least one empty voxel with an average of two or more neighbor voxels.

29. The method of any one of claims 25-28, wherein the information corresponding to the point cloud further corresponds to one or more voxels.

30. The method of any one of claims 25-29, wherein the subprocess is a normalizing flow (NF) subprocess.

31. The method of any one of claims 25-30, wherein the encoded bitstream comprises a signaling flag to indicate an architecture configuration.

32. An apparatus comprising: a processor; and a non-transitory computer-readable medium storing instructions operative, when executed by the processor, to cause the apparatus to perform the method of any one of claims 25 through 31 .

33. An apparatus comprising at least one processor configured to perform the method of any one of claims1-7, 9-15, 17-23, and 25-31.

34. An apparatus comprising a computer-readable medium storing instructions for causing one or more processors to perform the method of any one of claims 1-7, 9-15, 17-23, and 25-31.

35. An apparatus comprising at least one processor and at least one non-transitory computer-readable medium storing instructions for causing the at least one processor to perform the method of any one of claims 1-7, 9-15, 17-23, and 25-31.

36. A computer-readable medium storing a scene description file generated according to any one of claims1-7, 9-15, 17-23, and 25-31.

37. A signal including a scene description file generated according to any one of claims 1-7, 9-15, 17-23, and

Images