公开(公告)号：US20220245322A1

公开(公告)日：2022-08-04

申请号：US17163162

申请日：2021-01-29

Applicant: salesforce.com, inc.

Inventor： Jessica Lundin ,  Owen Winne Schoppe ,  Xing Han ,  Michael Reynolds Sollami ,  Brian J. Lonsdorf ,  Alan Martin Ross ,  David J. Woodward ,  Sonke Rohde

IPC: G06F40/103 ,  G06F40/284 ,  G06T11/60 ,  G06N3/04

Abstract: An online system generates a set of content item variations for a reference content item that include different styles of text for the content item. The different styles of text are generated by applying machine-learned style transfer models, for example, neural network based models to reference text of the reference content item. The text variations retain the textual content of the reference text but are synthesized with different styles. The online system can provide the content item variations to users on an online experimental platform to collect user interaction information that may indicate how users respond to different styles of text. The online system or the content providers can effectively target users with content items that include the style of text the users respond to based on the collected information.

公开(公告)号：US11379189B1

公开(公告)日：2022-07-05

申请号：US17354439

申请日：2021-06-22

Applicant: salesforce.com, inc.

Inventor： Owen Winne Schoppe ,  Sönke Rohde ,  Brian J. Lonsdorf ,  Jessica Lundin ,  David J. Woodward ,  Alan Martin Ross ,  Michael Sollami

IPC: G06F8/20 ,  G06F8/38 ,  G06N20/00 ,  G06F9/451

Abstract: Techniques are disclosed relating to automatically synthesizing user interface (UI) component instances. In disclosed techniques a computer system receives a set of existing UI elements and a set of design rules for the set of existing elements, where design rules in the set of design rules indicate one or more allowed states for respective UI elements in the set of existing UI elements. The one or more allowed states may correspond to one or more visual characteristics. Using the set of existing UI elements, the computer system may then automatically generate a plurality of UI component instances based on the set of design rules, where a respective UI component instance includes a first UI element in a first allowed state. The computer system may then train, using the plurality of UI component instances, a machine learning model operable to automatically generate UI designs.

公开(公告)号：US11074044B1

公开(公告)日：2021-07-27

申请号：US17147053

申请日：2021-01-12

Applicant: salesforce.com, inc.

Inventor： Owen Winne Schoppe ,  Sönke Rohde ,  Brian J. Lonsdorf ,  Jessica Lundin ,  David J. Woodward ,  Alan Martin Ross ,  Michael Sollami

IPC: G06F8/20 ,  G06N20/00 ,  G06F8/38 ,  G06F9/451

Abstract: Techniques are disclosed relating to automatically synthesizing user interface (UI) component instances. In disclosed techniques a computer system receives a set of existing UI elements and a set of design rules for the set of existing elements, where design rules in the set of design rules indicate one or more allowed states for respective UI elements in the set of existing UI elements. The one or more allowed states may correspond to one or more visual characteristics. Using the set of existing UI elements, the computer system may then automatically generate a plurality of UI component instances based on the set of design rules, where a respective UI component instance includes a first UI element in a first allowed state. The computer system may then train, using the plurality of UI component instances, a machine learning model operable to automatically generate UI designs.

公开(公告)号：US11798210B2

公开(公告)日：2023-10-24

申请号：US17116944

申请日：2020-12-09

Applicant: salesforce.com, inc.

Inventor： Jessica Lundin ,  Michael Reynolds Sollami ,  Alan Martin Ross ,  Brian J. Lonsdorf ,  David James Woodward ,  Owen Winne Schoppe ,  Sönke Rohde

IPC: G06T11/60 ,  G06T7/70 ,  G06N3/08 ,  G06F18/214

CPC classification number: G06T11/60 ,  G06F18/214 ,  G06N3/08 ,  G06T7/70 ,  G06T2207/10016 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2210/12

Abstract: Disclosed herein are system, method and computer readable storage medium for detecting space suitable for overlaying media content onto an image. The system receives a candidate image which may be an image or a video frame. The candidate image is then input into a neural network. The neural network may output coordinates and one or more dimensions representing one or more bounding boxes for inserting media content into the candidate image. The one or more bounding boxes may be transmitted with a request for a media content item to be displayed in a bounding box. In response to the request the media content item may be received, and the candidate image and the media content item overlaid on top of the candidate image within the bounding box may be displayed.

公开(公告)号：US11694018B2

公开(公告)日：2023-07-04

申请号：US17163162

申请日：2021-01-29

Applicant: salesforce.com, inc.

Inventor： Jessica Lundin ,  Owen Winne Schoppe ,  Xing Han ,  Michael Reynolds Sollami ,  Brian J. Lonsdorf ,  Alan Martin Ross ,  David J. Woodward ,  Sonke Rohde

IPC: G06F40/169 ,  G06F40/103 ,  G06N3/049 ,  G06T11/60 ,  G06F40/284

CPC classification number: G06F40/103 ,  G06F40/284 ,  G06N3/049 ,  G06T11/60

Abstract: An online system generates a set of content item variations for a reference content item that include different styles of text for the content item. The different styles of text are generated by applying machine-learned style transfer models, for example, neural network based models to reference text of the reference content item. The text variations retain the textual content of the reference text but are synthesized with different styles. The online system can provide the content item variations to users on an online experimental platform to collect user interaction information that may indicate how users respond to different styles of text. The online system or the content providers can effectively target users with content items that include the style of text the users respond to based on the collected information.

公开(公告)号：US20230177269A1

公开(公告)日：2023-06-08

申请号：US17545168

申请日：2021-12-08

Applicant: salesforce.com, inc.

Inventor： Jessica Lundin ,  Sönke Rohde ,  Owen Winne Schoppe ,  Michael Sollami ,  David Woodward ,  Brian Lonsdorf ,  Alan Martin Ross ,  Scott Bokma

IPC: G06F40/289 ,  G06N20/00

CPC classification number: G06F40/289 ,  G06N20/00

Abstract: Systems, devices, and techniques are disclosed for conversation topic extraction. Text of a communication channel may be received. The text of the communication channel may be divided into conversation documents based on conversation threads of the communication channel. Phrases of the text of the conversation documents may be tokenizes. Topic phrases for the conversation documents may be determined by assigning importance scores to the tokenized phrases using unsupervised topic extraction. The topic phrases may be the tokenized phrases with the highest importance scores.

公开(公告)号：US11409416B2

公开(公告)日：2022-08-09

申请号：US16941176

申请日：2020-07-28

Applicant: salesforce.com, inc.

Inventor： David James Woodward ,  Brian J. Lonsdorf ,  Owen Winne Schoppe ,  Alan Martin Ross ,  Jessica Lundin ,  Sönke Rohde

IPC: G06F3/0484 ,  G06F9/451 ,  G06F3/0481 ,  G06N20/00 ,  G06N5/04

Abstract: Disclosed herein are system, method, and computer program product embodiments for generating custom user interfaces (UIs) for completing a task. One embodiment operates by obtaining contextual information associated with a user and an application on a user device operated by the user, where the application includes a plurality of UI elements. Then, determining the user is attempting to complete a first task within the application based on the contextual information and a prediction model. The embodiment further operates by obtaining a minimum set of UI elements required for the first task. Further, the embodiment operates by transmitting a first custom UI including the minimum set of UI elements for the first task to the user device for display to the user.

公开(公告)号：US20210334666A1

公开(公告)日：2021-10-28

申请号：US16854913

申请日：2020-04-22

Applicant: salesforce.com, inc.

Inventor： Jessica Lundin ,  Owen Winne Schoppe ,  Alan Martin Ross ,  Brian J. Lonsdorf ,  David James Woodward ,  Sönke Rohde ,  Michael Reynolds Sollami ,  Chetan Ramaiah

IPC: G06N5/02 ,  G06N20/00 ,  G06F40/109 ,  G06F17/16

Abstract: A textual properties model is used to infer values for certain font properties of interest given certain text-related data, such as rendered text images. The model may be used for numerous purposes, such as aiding with document layout, identifying font families that are similar to a given font families, and generating new font families with specific desired properties. In some embodiments, the model is trained from a combination of synthetic data that is labeled with values for the font properties of interest, and partially-labeled data from existing “real-world” documents.

公开(公告)号：US11137985B2

公开(公告)日：2021-10-05

申请号：US16779215

申请日：2020-01-31

Applicant: salesforce.com, inc.

Inventor： Owen Winne Schoppe ,  Brian J. Lonsdorf ,  Alan Martin Ross ,  Sönke Rohde ,  David James Woodward ,  Jessica Lundin

IPC: G06F8/38 ,  G06T11/20 ,  G06T7/11 ,  G06F16/957

Abstract: Techniques are disclosed for automatically generating stencils for content of a user interface (UI) to be rendered. A computer system receives information specifying content of a user interface (UI) to be rendered. Based on this information, the computer system identifies one or more bounding regions of content within the UI, including analyzing metadata and a rendered version of the UI. The computer system then automatically generates, one or more UI stencils based on the identified bounding regions, that are displayable as progress indicators prior to rendering corresponding UI content. Once it has generated the stencils, the computer system stores information specifying the one or more UI stencils for use in the UI. Automatically generating stencils for UIs may advantageously reduce the time gap from development to deployment of these UIs while improving their perceived performance and, by extension, improving user experience.

公开(公告)号：US11868790B2

公开(公告)日：2024-01-09

申请号：US17649016

申请日：2022-01-26

Applicant: salesforce.com, inc.

Inventor： Michael Sollami ,  Sönke Rohde ,  Alan Martin Ross ,  David James Woodward ,  Jessica Lundin ,  Owen Winne Schoppe ,  Brian J. Lonsdorf ,  Aashish Jain

IPC: G06F9/451 ,  G06N3/08 ,  G06F9/54 ,  G06N3/045 ,  G06V10/762 ,  G06V10/771 ,  G06V10/82 ,  G06F3/04845 ,  G06N3/088 ,  G06F8/38 ,  G06N3/047 ,  G06N3/044 ,  G06N7/01 ,  G06V30/19 ,  G06F17/00

CPC classification number: G06F9/451 ,  G06F3/04845 ,  G06F9/547 ,  G06N3/045 ,  G06N3/08 ,  G06V10/763 ,  G06V10/771 ,  G06V10/82 ,  G06F8/38 ,  G06N3/044 ,  G06N3/047 ,  G06N3/088 ,  G06N7/01 ,  G06V30/19173

Abstract: Techniques are disclosed for automatically generating new content using a trained 1-to-N generative adversarial network (GAN) model. In disclosed techniques, a computer system receives, from a computing device, a request for newly-generated content, where the request includes current content. The computer system automatically generates, using the trained 1-to-N GAN model, N different versions of new content, where a given version of new content is automatically generated based on the current content and one of N different style codes, where the value of N is at least two. After generating the N different versions of new content, the computer system transmits them to the computing device. The disclosed techniques may advantageously automate a content generation process, thereby saving time and computing resources via execution of the 1-to-N GAN machine learning model.