公开(公告)号：US11900558B2

公开(公告)日：2024-02-13

申请号：US17520071

申请日：2021-11-05

申请人： Adobe Inc.

发明人： Marissa Ramirez de Chanlatte ,  Radomir Mech ,  Matheus Gadelha ,  Thibault Groueix

IPC分类号： G06T19/20 ,  G06N3/04 ,  G06T7/50 ,  G06F18/214 ,  G06T17/00

CPC分类号： G06T19/20 ,  G06F18/214 ,  G06N3/04 ,  G06T7/50 ,  G06T17/00 ,  G06T2219/2004

摘要： The present disclosure relates to systems, non-transitory computer-readable media, and methods that tune a 3D-object-reconstruction-machine-learning model to reconstruct 3D models of objects from real images using real images as training data. For instance, the disclosed systems can determine a depth map for a real two-dimensional (2D) image and then reconstruct a 3D model of a digital object in the real 2D image based on the depth map. By using a depth map for a real 2D image, the disclosed systems can generate reconstructed 3D models that better conform to the shape of digital objects in real images than existing systems and use such reconstructed 3D models to generate more realistic looking visual effects (e.g., shadows, relighting).

公开(公告)号：US20210201150A1

公开(公告)日：2021-07-01

申请号：US17204370

申请日：2021-03-17

申请人： Adobe Inc.

发明人： Zhe Lin ,  Xiaohui Shen ,  Radomir Mech ,  Jian Ren

IPC分类号： G06N3/08 ,  G06K9/62 ,  G06K9/00 ,  G06F16/783 ,  G06N3/04

摘要： Various embodiments describe frame selection based on training and using a neural network. In an example, the neural network is a convolutional neural network trained with training pairs. Each training pair includes two training frames from a frame collection. The loss function relies on the estimated quality difference between the two training frames. Further, the definition of the loss function varies based on the actual quality difference between these two frames. In a further example, the neural network is trained by incorporating facial heatmaps generated from the training frames and facial quality scores of faces detected in the training frames. In addition, the training involves using a feature mean that represents an average of the features of the training frames belonging to the same frame collection. Once the neural network is trained, a frame collection is input thereto and a frame is selected based on generated quality scores.

公开(公告)号：US10810721B2

公开(公告)日：2020-10-20

申请号：US15458826

申请日：2017-03-14

申请人： Adobe Inc.

发明人： Radomir Mech ,  Ning Yu ,  Xiaohui Shen ,  Zhe Lin

IPC分类号： G06T7/00 ,  G06N3/08 ,  G06N3/04

摘要： Digital image defect identification and correction techniques are described. In one example, a digital medium environment is configured to identify and correct a digital image defect through identification of a defect in a digital image using machine learning. The identification includes generating a plurality of defect type scores using a plurality of defect type identification models, as part of machine learning, for a plurality of different defect types and determining the digital image includes the defect based on the generated plurality of defect type scores. A correction is generated for the identified defect and the digital image is output as included the generated correction.

公开(公告)号：US10783431B2

公开(公告)日：2020-09-22

申请号：US14938752

申请日：2015-11-11

申请人： Adobe Inc.

发明人： Zeke Koch ,  Gavin Stuart Peter Miller ,  Jonathan W. Brandt ,  Nathan A. Carr ,  Radomir Mech ,  Walter Wei-Tuh Chang ,  Scott D. Cohen ,  Hailin Jin

IPC分类号： G06F17/30 ,  G06N3/08 ,  G06F16/583 ,  G06N20/00 ,  G06N5/02

摘要： Image search techniques and systems involving emotions are described. In one or more implementations, a digital medium environment of a content sharing service is described for image search result configuration and control based on a search request that indicates an emotion. The search request is received that includes one or more keywords and specifies an emotion. Images are located that are available for licensing by matching one or more tags associated with the image with the one or more keywords and as corresponding to the emotion. The emotion of the images is identified using one or more models that are trained using machine learning based at least in part on training images having tagged emotions. Output is controlled of a search result having one or more representations of the images that are selectable to license respective images from the content sharing service.

公开(公告)号：US10776671B2

公开(公告)日：2020-09-15

申请号：US15989436

申请日：2018-05-25

申请人： Adobe Inc.

发明人： Zhe Lin ,  Xiaohui Shen ,  Shanghang Zhang ,  Radomir Mech

IPC分类号： G06K9/62 ,  G06K9/46 ,  G06T7/00 ,  G06N3/08

摘要： Techniques are disclosed for blur classification. The techniques utilize an image content feature map, a blur map, and an attention map, thereby combining low-level blur estimation with a high-level understanding of important image content in order to perform blur classification. The techniques allow for programmatically determining if blur exists in an image, and determining what type of blur it is (e.g., high blur, low blur, middle or neutral blur, or no blur). According to one example embodiment, if blur is detected, an estimate of spatially-varying blur amounts is performed and blur desirability is categorized in terms of image quality.

公开(公告)号：US10516830B2

公开(公告)日：2019-12-24

申请号：US15730614

申请日：2017-10-11

申请人： Adobe Inc.

发明人： Jianming Zhang ,  Zijun Wei ,  Zhe Lin ,  Xiaohui Shen ,  Radomir Mech

IPC分类号： H04N5/232 ,  G06N3/08

摘要： Various embodiments describe facilitating real-time crops on an image. In an example, an image processing application executed on a device receives image data corresponding to a field of view of a camera of the device. The image processing application renders a major view on a display of the device in a preview mode. The major view presents a previewed image based on the image data. The image processing application receives a composition score of a cropped image from a deep-learning system. The image processing application renders a sub-view presenting the cropped image based on the composition score in a preview mode. Based on a user interaction, the image processing application renders the cropped image in the major view with the sub-view in the preview mode.

公开(公告)号：US20190213474A1

公开(公告)日：2019-07-11

申请号：US15866129

申请日：2018-01-09

申请人： Adobe Inc.

发明人： Zhe Lin ,  Xiaohui Shen ,  Radomir Mech ,  Jian Ren

IPC分类号： G06N3/08 ,  G06K9/62 ,  G06F17/30 ,  G06K9/00

CPC分类号： G06N3/08 ,  G06F16/784 ,  G06K9/00221 ,  G06K9/00744 ,  G06K9/6256

摘要： Various embodiments describe frame selection based on training and using a neural network. In an example, the neural network is a convolutional neural network trained with training pairs. Each training pair includes two training frames from a frame collection. The loss function relies on the estimated quality difference between the two training frames. Further, the definition of the loss function varies based on the actual quality difference between these two frames. In a further example, the neural network is trained by incorporating facial heatmaps generated from the training frames and facial quality scores of faces detected in the training frames. In addition, the training involves using a feature mean that represents an average of the features of the training frames belonging to the same frame collection. Once the neural network is trained, a frame collection is input thereto and a frame is selected based on generated quality scores.

公开(公告)号：US10346951B2

公开(公告)日：2019-07-09

申请号：US15448138

申请日：2017-03-02

申请人： Adobe Inc.

发明人： Zhe Lin ,  Radomir Mech ,  Xiaohui Shen ,  Brian L. Price ,  Jianming Zhang ,  Anant Gilra ,  Jen-Chan Jeff Chien

IPC分类号： G06K9/62 ,  G06T3/40 ,  G06K9/46 ,  G06T3/00 ,  G06T11/60

摘要： Image cropping suggestion using multiple saliency maps is described. In one or more implementations, component scores, indicative of visual characteristics established for visually-pleasing croppings, are computed for candidate image croppings using multiple different saliency maps. The visual characteristics on which a candidate image cropping is scored may be indicative of its composition quality, an extent to which it preserves content appearing in the scene, and a simplicity of its boundary. Based on the component scores, the croppings may be ranked with regard to each of the visual characteristics. The rankings may be used to cluster the candidate croppings into groups of similar croppings, such that croppings in a group are different by less than a threshold amount and croppings in different groups are different by at least the threshold amount. Based on the clustering, croppings may then be chosen, e.g., to present them to a user for selection.

公开(公告)号：US10307969B2

公开(公告)日：2019-06-04

申请号：US15334527

申请日：2016-10-26

申请人： Adobe Inc.

发明人： Nathan A. Carr ,  Tim Christopher Reiner ,  Gavin Stuart Peter Miller ,  Radomir Mech

IPC分类号： B29C41/02 ,  B29C41/52 ,  B29C67/00 ,  B29C64/118 ,  B33Y10/00 ,  B33Y30/00 ,  B33Y50/02 ,  B29C64/106 ,  B29C64/112 ,  B29C64/20 ,  B29C64/386 ,  B29K105/00

摘要： This document describes techniques and apparatuses for 3D printing with small geometric offsets to affect surface characteristics. These techniques are capable of enabling fused-deposition printers to create 3D objects having desired surface characteristics, such as particular colors, images and image resolutions, textures, and luminosities. In some cases, the techniques do so using a single filament head with a single filament material. In some other cases, the techniques do so using multiple heads each with different filaments, though the techniques can forgo many switches between these heads. Each printing layer may use even a single filament from one head, thereby enabling surface characteristics while reducing starts and stops for filaments heads, which enables fewer artifacts or increases printing speed.

公开(公告)号：US20190164342A1

公开(公告)日：2019-05-30

申请号：US15825959

申请日：2017-11-29

申请人： Adobe Inc.

发明人： Vojtech Krs ,  Radomir Mech ,  Nathan A. Carr

IPC分类号： G06T17/20 ,  G06T17/10

摘要： Techniques are disclosed for generation of 3D structures. A methodology implementing the techniques according to an embodiment includes initializing systems configured to provide rules that specify edge connections between vertices and parametric properties of the vertices. The rules are applied to an initial set of vertices to generate 3D graphs for each of these vertex-rule-graph (VRG) systems. The initial set of vertices is associated with provided interaction surfaces of a 3D model. Skeleton geometries are generated for the 3D graphs, and an associated objective function is calculated. The objective function is configured to evaluate the fitness of the skeleton geometries based on given geometric and functional constraints. A 3D structure is generated through an iterative application of genetic programming techniques applied to the VRG systems to minimize the objective function. Receiving updated constraints and interaction surfaces, for incorporation in the iterative process.