# US 8,787,659 B2 · Automatic adaptation to image processing pipeline > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,787,659 B2 - Document ID: US8787659B2 - Status: Granted - Kind: B2 - Application number: 13/958,469 - Filed: 2013-08-02 - Published: 2013-11-28 - Granted: 2014-07-22 - Earliest claimed: 2011-09-02 - Assignees: Adobe Systems Incorporated - Applicants: Adobe Systems Incorporated - Inventors: Sylvain P. Paris; Jen-Chan Chien; Vladimir L. Bychkovsky - Web page: https://patentgrep.com/patent/US8787659B2 ## CPC - G06V 20/00: Scenes; Scene-specific elements (control of digital cameras H04N23/60) - G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting - G06T 1/20: Processor architectures; Processor configuration, e.g. pipelining - G06T 5/60: using machine learning, e.g. neural networks - G06F 16/583: using metadata automatically derived from the content - G06T 2207/20081: Training; Learning - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract Techniques are disclosed relating to generating generic labels, translating generic labels to image pipeline-specific labels, and automatically adjusting images. In one embodiment, generic labels may be generated. Generic algorithm parameters may be generated based on training a regression algorithm with the generic labels. The generic labels may be translated to pipeline-specific labels, which may be usable to automatically adjust an image. ## Claims ### Claim 1 A method comprising: generating a plurality of generic labels by a computing device for a plurality of image pairs in which: each of the image pairs includes a first image and a corresponding adjusted image that is an adjusted version of the first image; and each generic label relates one or more parameters of one of the first images of one of the image pairs to one or more parameters of the corresponding adjusted image of the one image pair; and generating a plurality of generic algorithm parameters by the computing device using the plurality of generic labels, the plurality of generic algorithm parameters configured to be used by an image processing pipeline to translate the generic labels into pipeline-specific labels to be used to automatically adjust a new image in which the generic labels, the generic algorithm parameters, a translation to the image processing pipeline, and a translation to a different image processing pipeline are configured to be provided to the image processing pipeline and the different image processing pipeline, each of the image processing pipeline and the different image processing pipeline configured to translate the generic labels into pipeline-specific labels based on the respective translation. ### Claim 2 (depends on 1) A method of claim 1, wherein the generating of the plurality of generic algorithm parameters involves use of one or more regression algorithms. ### Claim 3 (depends on 1) A method of claim 1, further comprising storing the generic algorithm parameters at different degrees of fidelity. ### Claim 4 (depends on 1) A method of claim 1, wherein at least one of the pipeline-specific labels of the image processing pipeline are different than a corresponding translated one of the pipeline-specific labels of the different image processing pipeline. ### Claim 5 (depends on 1) A method of claim 1, wherein the generating of the plurality of generic labels includes performing principal component analysis on the plurality of image pairs. ### Claim 6 (depends on 1) A method of claim 1, wherein the image processing pipeline is further configured to automatically adjust the new image based on the pipeline-specific labels and the generic algorithm parameters. ### Claim 7 (depends on 1) A method of claim 1, wherein the generic labels represent a global tonal adjustment. ### Claim 8 A computing device comprising one or more modules implemented at least in partly by hardware and configured to perform operations comprising: generating a plurality of generic labels for a plurality of image pairs in which: each of the image pairs includes a first image and a corresponding adjusted image that is an adjusted version of the first image; and each generic label relates one or more parameters of one of the first images of one of the image pairs to one or more parameters of the corresponding adjusted image of the one image pair; and generating a plurality of generic algorithm parameters using the plurality of generic labels, the plurality of generic algorithm parameters configured to be used by an image processing pipeline to translate the generic labels into pipeline-specific labels to be used to automatically adjust a new image in which the generic labels, the generic algorithm parameters, a translation to the image processing pipeline, and a translation to a different image processing pipeline are configured to be provided to the image processing pipeline and the different image processing pipeline, each of the image processing pipeline and the different image processing pipeline configured to translate the generic labels into pipeline-specific labels based on the respective translation. ### Claim 9 (depends on 8) A computing device of claim 8, wherein the generating of the plurality of generic algorithm parameters involves use of one or more regression algorithms. ### Claim 10 (depends on 8) A computing device of claim 8, wherein the one or more modules are further configured to store the generic algorithm parameters at different degrees of fidelity. ### Claim 11 (depends on 8) A computing device of claim 8, wherein at least one of the pipeline-specific labels of the image processing pipeline are different than a corresponding translated one of the pipeline-specific labels of the different image processing pipeline. ### Claim 12 (depends on 8) A computing device of claim 8, wherein the generating of the plurality of generic labels includes performing principal component analysis on the plurality of image pairs. ### Claim 13 (depends on 8) A computing device of claim 8, wherein the image processing pipeline is further configured to automatically adjust the new image based on the pipeline-specific labels and the generic algorithm parameters. ### Claim 14 (depends on 8) A computing device of claim 8, wherein the generic labels represent a global tonal adjustment. ### Claim 15 One or more computer-readable storage media that are non-transitory and comprising instructions that are stored thereon that, responsive to execution by a computing device, cause the computing device to perform operations comprising: generating a plurality of generic labels for a plurality of image pairs in which: each of the image pairs includes a first image and a corresponding adjusted image that is an adjusted version of the first image; and each generic label relates one or more parameters of one of the first images of one of the image pairs to one or more parameters of the corresponding adjusted image of the one image pair; and generating a plurality of generic algorithm parameters using the plurality of generic labels, the plurality of generic algorithm parameters configured to be used by an image processing pipeline to translate the generic labels into pipeline-specific labels to be used to automatically adjust a new image in which the generic labels, the generic algorithm parameters, a translation to the image processing pipeline, and a translation to a different image processing pipeline are configured to be provided to the image processing pipeline and the different image processing pipeline, each of the image processing pipeline and the different image processing pipeline configured to translate the generic labels into pipeline-specific labels based on the respective translation. ### Claim 16 (depends on 15) One or more computer-readable storage media as described in claim 15, wherein the generating of the plurality of generic algorithm parameters involves use of one or more regression algorithms. ### Claim 17 (depends on 15) One or more computer-readable storage media as described in claim 15, further comprising storing the generic algorithm parameters at different degrees of fidelity. ### Claim 18 (depends on 15) One or more computer-readable storage media as described in claim 15, wherein at least one of the pipeline-specific labels of the image processing pipeline are different than a corresponding translated one of the pipeline-specific labels of the different image processing pipeline. ### Claim 19 (depends on 15) One or more computer-readable storage media as described in claim 15, wherein the generating of the plurality of generic labels includes performing principal component analysis on the plurality of image pairs. ### Claim 20 (depends on 15) One or more computer-readable storage media as described in claim 15, wherein the image processing pipeline is further configured to automatically adjust the new image based on the pipeline-specific labels and the generic algorithm parameters.