专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

51. 发明专利

AU2019216594B2 FAST MOTION-BASED VIDEO INPAINTING 未知
公开(公告)号：AU2019216594B2
公开(公告)日：2021-08-05
申请号：AU2019216594
申请日：2019-08-12
申请人： ADOBE INC
发明人： OXHOLM GEOFFREY , SHEIKH RAMIZ , WANG OLIVER , LUKAC MICHAL , SHECHTMAN ELYA , WALKER SETH , NELSON JOHN
IPC分类号： G06T5/00 , G06T5/50
摘要： Abstract Certain aspects involve video inpainting via confidence-weighted motion estimation. For instance, a video editor accesses video content having a target region to be modified in one or more video frames. The video editor computes a motion for a boundary of the target region. The video editor interpolates, from the boundary motion, a target motion of a target pixel within the target region. In the interpolation, confidence values assigned to boundary pixels control how the motion of these pixels contributes to the interpolated target motion. A confidence value is computed based on a difference between forward and reverse motion with respect to a particular boundary pixel, a texture in a region that includes the particular boundary pixel, or a combination thereof. The video editor modifies the target region in the video by updating color data of the target pixel to correspond to the target motion interpolated from the boundary motion. CfI CDa 10 * I _i -~CI Ia, -c w ~ ~CN >LL Cf) LC - - - I

52. 发明专利

AU2019200746B2 METHOD TO GENERATE SUMMARIES TUNED TO TOPICS OF INTEREST OF READERS 未知
公开(公告)号：AU2019200746B2
公开(公告)日：2021-07-29
申请号：AU2019200746
申请日：2019-02-04
申请人： ADOBE INC
发明人： KRISHNA KUNDAN , SRINIVASAN BALAJI VASAN
IPC分类号： G06F16/34 , G06F40/00 , G06N3/02
摘要： A word generation model obtains textual content and a requested topic of interest, and generates a targeted summary of the textual content tuned to the topic of interest. To do so, a topic-aware encoding model encodes the textual content with a topic label corresponding to the topic of interest to generate topic-aware encoded text. A word generation model selects a next word for the topic-based summary from the topic-aware encoded text. The word generation model is trained to generate topic-based summaries using machine learning on training data including a multitude of documents, a respective summary of each document, and a respective topic of each summary. Feedback of the selected next word is provided to the word generation model. The feedback causes the word generation model to select subsequent words for the topic based summary based on the feedback of the next selected word. Service Provider System 102 Request Topic-Based Summary Training Module Module 118 124 Cxtuat Word Generation Model Articficially 112 120 1 Generated -- Training Data Topic of Topic-Aware Encoding 126 Interest Model 122 Topic-Based Summary Network Client Device 104 Communication Module 108

53. 发明专利

AU2018250372B2 METHOD TO CONSTRUCT CONTENT BASED ON A CONTENT REPOSITORY 未知
公开(公告)号：AU2018250372B2
公开(公告)日：2021-07-29
申请号：AU2018250372
申请日：2018-10-16
申请人： ADOBE INC
发明人： SRINIVASAN BALAJI VASAN , MANERIKER PRANAV RAVINDRA , MODANI NATWAR , KRISHNA KUNDAN
IPC分类号： G06F16/31 , G06F16/33 , G06F16/338 , G06F40/151
摘要： Embodiments of the present invention provide systems, methods, and computer storage media directed to facilitating corpus-based content generation, in particular, using graph-based multi-sentence compression to generate a final content output. In one embodiment, pre-existing source content is identified and retrieved from a corpus. The source content is then parsed into sentence tokens, mapped and weighted. The sentence tokens are further parsed into word tokens and weighted. The mapped word tokens are then compressed into candidate sentences to be used in a final content. The final content is assembled using ranked candidate sentences, such that the final content is organized to reduce information redundancy and optimize content cohesion. (0 Cw Io 0

54. 发明专利

AU2018206708B2 PROJECT FACES 未知
公开(公告)号：AU2018206708B2
公开(公告)日：2021-07-22
申请号：AU2018206708
申请日：2018-07-17
申请人： ADOBE INC
发明人： SINN RICHARD , DONAHUE THOMAS , YOUNG ALAN
IPC分类号： G06F17/00
摘要： Abstract A typeface processing application accesses a character skeleton, which has control points and curves defined by control points, and a character outline surrounding the character skeleton. The typeface processing application computes, for a design parameter of a design application, a graphics control dataset based on a particular control point. Doing so involves identifying a pair of control point positions and a pair of expansions of the character outline, where each position or expansion pair corresponds to a pair of design parameter values. The typeface processing application generates the graphics control dataset, which includes intermediate positions of the control point between the pair of positions and intermediate expansions of the character outline between the pair of expansions. The typeface processing application outputs the graphics control dataset to the design application, which can display, responsive to a selection of a design parameter value, modified character designs generated using the graphics control dataset.

55. 发明专利

AU2018250521B2 BLUE NOISE DITHERED FILTERING 未知
公开(公告)号：AU2018250521B2
公开(公告)日：2021-07-08
申请号：AU2018250521
申请日：2018-10-21
申请人： ADOBE INC
发明人： WEI LI-YI , SUN XIN
IPC分类号： G06T3/40
摘要： Abstract In some embodiments, a computing device uses a blue noise sampling operation to identify source pixels from an input image defining respective pixel sets. Each pixel set is associated with a respective weight matrix for a down-scaling operation. The blue noise sampling operation causes an overlap region between first and second pixel sets. The computing device assigns an overlap pixel in the overlap region to the first weight matrix based on the overlap pixel being closer to the first source pixel. The computing device modifies the second weight matrix to exclude the overlap pixel from a portion of the down-scaling operation involving the second weight matrix. The computing device performs the down-scaling operation on the input image by combining the first pixel set into a first target pixel with the first weight matrix and combining the second pixel set into a second target with the modified second weight matrix. SEARCH LAYER F- t t REPRESENTATION REPRESENTATION GENERATOR GENERATOR QUERY SEARCH OBJECT MEDIUM

56. 发明专利

AU2017202712B2 ILLUMINATION-GUIDED EXAMPLE-BASED STYLIZATION OF 3D SCENES 未知
公开(公告)号：AU2017202712B2
公开(公告)日：2021-07-08
申请号：AU2017202712
申请日：2017-04-25
申请人： ADOBE INC , CZECH TECHNICAL UNIV IN PRAGUE
发明人： SHECHTMAN ELYA , ASENTE PAUL J , FISER JAKUB , LU JINGWAN , SÝKORA DANIEL , JAMRIŠKA ONDŘEJ , LUKÁČ MICHAL
IPC分类号： G06T15/50
摘要： A method is provided in a digital medium environment to control patch-usage in image synthesis based on matching errors associated with potential source-to-target patch assignments between a source image and a target image. The method comprising determining an error budget using a curve fitted to a sorted set of the matching errors, the error budget being usable to identify feasible patch assignments from the potential source-to-target patch assignments. Assigning source patches from the source image to target patches in the target image using the error budget and uniform patch usage enforcement. Assigning at least one of the source patches to at least one additional target patch in the target image based on the error budget. Synthesizing an image based on the source patches assigned to the target patches. Inventors: Fier et al. Controlling Patch Usage in Image Synthesis 100 / Service Provider 102 Service Manager Module 108 Image Service 110 Image Synthesis Module 112 Error Control Module 114 Network Computing Device 104 130 Processing System 120 \ Communication 7 Module 122 / Image Processing B Module 126 [ Memory 124 e._________ 136 128--

57. 发明专利

AU2016225778B2 DEHAZING PHOTOS AND VIDEOS USING VISUAL ARTIFACT SUPPRESSION 未知
公开(公告)号：AU2016225778B2
公开(公告)日：2021-06-17
申请号：AU2016225778
申请日：2016-09-05
申请人： ADOBE INC
发明人： CHEN CHEN , WANG JUE
IPC分类号： G06T5/00
摘要： FTHE DISCLOSURE Methods and systems for dehazing images with increased accuracy and reduced error enhancement. In particular, one or more embodiments estimate a transmission map representing an amount of unscattered light reflected from objects in an input image. One or more embodiments refine the transmission map to obtain transmission information consistent with a depth of the objects in the input image. One or more embodiments also determine a radiance gradient for the input image. One or more embodiments generate an output image from the input image by removing haze based on the refined transmission map and preventing error enhancement based on the determined radiance gradient. Attorney DocketNo. 20030.51 49 Parent Application

58. 发明专利

AU2017204531B2 SMART AND AUTOMATIC SELECTION OF SAME SHAPED OBJECTS IN IMAGES 未知
公开(公告)号：AU2017204531B2
公开(公告)日：2021-05-27
申请号：AU2017204531
申请日：2017-07-01
申请人： ADOBE INC
发明人： GUPTA SOURABH , BEDI AJAY
IPC分类号： G06K9/46 , G06T7/12 , G06T7/13 , G06T7/136
摘要： Abstract A method for automatically selecting similar objects for modification within images stored in a computing device, comprising: receiving, by a processor of a computing device, input via an input device of the computing device specifying a first selection of a first object in an image displayed on a user interface of the computing device; generating, by the processor, a first set of feature descriptors that describe a shape formed by edges of the first object; identifying, by the processor, a plurality of other edges in the remaining portion of the image and generating a second set of feature descriptors describing shapes formed by the other edges; determining, by the processor, one or more edge objects in the image that are similar in shape to the first object. , 900 Receive input specifying a selection of a first object displayed in an image NO Identify other similar objects? Generate object feature descriptors describing the shape formed by the edges of the first object Identify other edges in the image and generate image feature descriptors describing the shapes formed by the other edges Determine one or more edge objects in the image that are similar in shape to the first object Modify pixels in the image of the first object and each similar edge object based on a specified function

59. 发明专利

AU2016225946B2 STYLE TRANSFER FOR HEADSHOT PORTRAITS 未知
公开(公告)号：AU2016225946B2
公开(公告)日：2021-05-20
申请号：AU2016225946
申请日：2016-09-12
申请人： ADOBE INC
发明人： BAILEY ROBERT , PARIS SYLVAIN , BARZEL RONEN , SACHS IAN
IPC分类号： G06T5/50
摘要： STYLE TRANSFER FOR HEADSHOT PORTRAITS [0039] Techniques are disclosed for automatically transferring a style of at least two reference images to an input image. The resulting transformation of the input image matches the visual styles of the reference images without changing the identity of the subject of the input image. Each image is decomposed into levels of detail with corresponding energy levels and a residual. A style transfer operation is performed at each energy level and residual using the reference image that most closely matches the input image at each energy level. The transformations of each level of detail and the residual of the input image are aggregated to generate an output image having the styles of the reference images. In some cases, the transformations are performed on the foreground of the input image, and the background can be transformed by an amount that is proportional to the aggregated transformations of the foreground. Docket # ADO.P5730US 31 nput Image Reference I mages Computing Device 150 Style Transfer Application 120 File Server Pose Calculation and Image Pre Decomposition computed Module Reference 12S mage Data Detail,1316 Energy1316 Levels and /Output Residual / Image / Network Signature Image Matching Calculation and Style Transfer Module Module 124 126Server Peg Application 120 Enegy .2 tl Tase Level and Residual Signatures/

60. 发明专利

AU2016277656B2 CONTEXT-BASED RETRIEVAL AND RECOMMENDATION IN THE DOCUMENT CLOUD 未知
公开(公告)号：AU2016277656B2
公开(公告)日：2021-05-13
申请号：AU2016277656
申请日：2016-12-22
申请人： ADOBE INC
发明人： KOH EUNYEE , GOLLUB TIM , LIPKA NEDIM
IPC分类号： G06F17/00 , G06F15/00
摘要： A computer-implemented method for organizing a plurality of files stored in an electronic file system, the method comprising generating a plurality of similarity scores for corresponding pairs of files of the plurality of files stored in a first directory of the electronic file system; identifying at least two similar files based on the generated similarity scores; creating a second directory and dynamically organizing the plurality of files by storing the at least two similar files in the second directory so that the number of files stored in the first directory is reduced, and files of the plurality of files are grouped together in one of the first directory or the second directory based on similarities in content indicated by the generated similarity scores. Store plurality of files in a first directory, the number of files in the plurality equal to a threshold Receive instruction to store additional file at first directory so that the number of files of the plurality plus the additional file in the first directory exceeds the threshold Generate vector representations of each file of the plurality and the additional file Responsive to the number of files exceeding the threshold Generate a plurality of similarity scores for corresponding pairs of files of the plurality of files stored in a first directory of the electronic file system Identify at least two similar files based on the generated similarity scores Create second directory Dynamically organize the plurality of files by storing the at least two similar files in the second directory so that the number of files stored in the first directory is reduced, and files of the plurality of files are grouped together in one of the first directory or the second directory based on similarities in content indicated by the generated similarity scores

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式