# US 2021/0241431 A1 · Apparatus and method for efficient distributed denoising of a graphics frame > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0241431 A1 - Document ID: US20210241431A1 - Status: Published application - Kind: A1 - Application number: 17/171,925 - Filed: 2021-02-09 - Published: 2021-08-05 - Earliest claimed: 2018-12-21 - Assignees: Intel Corporation - Applicants: Intel Corporation - Inventors: Carson BROWNLEE; Ingo WALD; Attila AFRA; Johannes GUENTHER; Jefferson AMSTUTZ; Carsten BENTHIN - Web page: https://patentgrep.com/patent/US20210241431A1 ## CPC - G06T 5/70: Denoising; Smoothing - G06T 1/20: Processor architectures; Processor configuration, e.g. pipelining - G06T 5/20: using local operators - G06T 5/60: using machine learning, e.g. neural networks - G06T 9/00: Image coding (bandwidth or redundancy reduction for static pictures H04N1/41; coding or decoding of static colour picture signals H04N1/64; methods or arrangements for coding, decoding, compressing or decompressing digital video signals H04N19/00) - G06T 15/06: Ray-tracing - G06T 2207/20081: Training; Learning - G06T 2207/20084: Artificial neural networks [ANN] - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract Apparatus and method for denoising of images generated by a rendering engine such as a ray tracing engine. For example, one embodiment of a system or apparatus comprises: A system comprising: a plurality of nodes to perform ray tracing operations; a dispatcher node to dispatch graphics work to the plurality of nodes, each node to perform ray tracing to render a region of an image frame; at least a first node of the plurality comprising: a ray-tracing renderer to perform ray tracing to render a first region of the image frame; and a denoiser to perform denoising of the first region using a combination of data associated with the first region and data associated with a region outside of the first region, at least some of the data associated with the region outside of the first region to be retrieved from at least one other node. ## Claims ### Claim 1 A node to perform ray tracing operations, comprising: a ray tracing renderer to perform ray tracing to render a first region of an image frame; a denoiser to perform denoising of the first region using a combination of data associated with the first region and data associated with a region outside of the first region, at least some of the data associated with the region outside of the first region to be retrieved from another node; and a compressor/decompressor to decompress the data obtained from the other node prior to being used to denoise the first region. ### Claim 2 (depends on 1) The node of claim 1, further comprising a network interface to couple the node with the other node through a high-speed network, a bus, or an interconnect. ### Claim 3 (depends on 2) The node of claim 2, wherein using the data obtained from the other node to perform denoising is based on a determination that a network connectivity provided by the network interface is faster than a threshold. ### Claim 4 (depends on 1) The node of claim 1, wherein the denoiser is to perform the denoising of the first region on one of a plurality of levels of quality, and wherein each level of quality corresponds to a different size of the region outside of the first region. ### Claim 5 (depends on 4) The node of claim 4, wherein prior to obtaining the data from the other node, the node is to determine that for a specified level of quality, the node does not have the data from the other node for the region outside of the first region corresponding to the specified level of quality. ### Claim 6 (depends on 1) The node of claim 1, wherein the denoiser includes a machine learning engine to evaluate the data associated with the first region and the region outside of the first region to perform denoising of the first region. ### Claim 7 (depends on 6) The node of claim 6, wherein the machine learning engine comprises convolutional neural network circuitry and logic to evaluate the data associated with the first region and the region outside of the first region based on a set of trained weights. ### Claim 8 (depends on 7) The node of claim 7, wherein the node is to receive at least a portion of a set of trained weights for machine learning to evaluate the data associated with the first region and the region outside of the first region. ### Claim 9 (depends on 8) The node of claim 8, further comprising: training circuitry and/or logic to perform runtime training using at least a portion of the data associated with the first region and to responsively perform an update to the set of trained weights and communicate the update to the other node. ### Claim 10 A method comprising: performing ray tracing operations on a node to render a first region of an image frame; requesting data associated with a region outside of the first region from another node; decompressing the data obtained from the other node; and denoising the first region using a combination of data associated with the first region and the data associated with the region outside of the first region. ### Claim 11 (depends on 10) The method of claim 10, wherein the data is obtained through a network interface that couples the node with the other node through a high-speed network, a bus, or an interconnect. ### Claim 12 (depends on 11) The method of claim 11, wherein using the data obtained from the other node to perform denoising is based on a determination that a network connectivity provided by the network interface is faster than a threshold. ### Claim 13 (depends on 10) The method of claim 10, wherein the denoising of the first region is performed on one of a plurality of levels of quality, and wherein each level of quality corresponds to a different size of the region outside of the first region. ### Claim 14 (depends on 13) The method of claim 13, wherein prior to obtaining the data from the other node, the method further comprises determining that for a specified level of quality, the node does not have the data from the other node for the region outside of the first region corresponding to the specified level of quality ### Claim 15 (depends on 10) The method of claim 10, wherein the denoising of the first region comprises performing machine learning to evaluate the data associated with the first region and the region outside of the first region to perform denoising of the first region. ### Claim 16 A non-transitory computer machine-readable medium having program code stored thereon which, when executed by a machine, is capable of causing the machine to perform the operations of: performing ray tracing operations on a node to render a first region of an image frame; requesting data associated with a region outside of the first region from another node; decompressing the data obtained from the other node; and denoising the first region using a combination of data associated with the first region and the data associated with the region outside of the first region. ### Claim 17 (depends on 16) The non-transitory computer machine-readable medium of claim 16, wherein the data is obtained through a network interface that couples the node with the other node through a high-speed network, a bus, or an interconnect. ### Claim 18 (depends on 17) The non-transitory computer machine-readable medium of claim 17, wherein using the data obtained from the other node to perform denoising is based on a determination that a network connectivity provided by the network interface is faster than a threshold. ### Claim 19 (depends on 16) The non-transitory computer machine-readable medium of claim 16, wherein the denoising of the first region is performed on one of a plurality of levels of quality, and wherein each level of quality corresponds to a different size of the region outside of the first region. ### Claim 20 (depends on 16) The non-transitory computer machine-readable medium of claim 16, wherein the denoising of the first region comprises performing machine learning to evaluate the data associated with the first region and the region outside of the first region to perform denoising of the first region.