Patents by Inventor Shaomei Wu
Shaomei Wu has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20220398822Abstract: The present application is at least directed to a system including a processor and non-transitory memory including computer-executable instructions, which when executed by the processor, perform receiving, via a user operating the system, a selection of a mode for recognizing objects. The processor is also configured to execute the instructions of causing a camera operably coupled to the system to operate in the selected mode for recognizing objects. The processor is further configured to execute the instructions of receiving, via the user operating the system, an image of a selected object. The processor is even further configured to execute the instructions of evaluating, via a trained machine learning model, one or more attributes of the selected object. The processor is yet further configured to execute the instructions of generating an image description based on at least a subset of the evaluated one or more attributes.Type: ApplicationFiled: April 28, 2022Publication date: December 15, 2022Inventors: Shaomei Wu, Tatiana Iskandar, Dennis Stewart William Tansley, Jerry Guanhua Qian, Isaac Robinson, Jerry L. Robinson, Madhavi Marigold Muppala
-
Patent number: 10489507Abstract: In one embodiment, a method includes identifying a plurality of dyslexic users on an online social network. The plurality of dyslexic users may be identified based on content objects posted by these users over a particular time period, where the content objects may include one or more of word-level errors or sentence-level errors. A machine-learning model may be trained for text correction using a corpus of social network data, which may include at least the content objects with one or more of word-level errors or sentence-level errors, and a corresponding set of corrected content objects. A text string including one or more errors may be received from a client system associated with a first user. The text string may be transformed into a vector representation using an encoder of the machine-learning model. A corrected text string may be generated from the vector representation using a decoder of the machine-learning model.Type: GrantFiled: January 2, 2018Date of Patent: November 26, 2019Assignee: Facebook, Inc.Inventors: Xian Li, Irina-Elena Veliche, Debnil Sur, Shaomei Wu, Amit Bahl, Juan Miguel Pino
-
Patent number: 10360466Abstract: Systems, methods, and non-transitory computer-readable media can receive an image. One or more concepts depicted in the image are identified based on machine learning techniques. The one or more concepts are filtered based on filtering criteria to identify one or more selected concepts. An image description is generated comprising the one or more selected concepts.Type: GrantFiled: December 27, 2016Date of Patent: July 23, 2019Assignee: Facebook, Inc.Inventors: Shaomei Wu, Lada Ariana Adamic, Jeffrey C. Wieland, Omid Farivar, Hermes Germi Pique Corchs, Matt King, Brett Alden Lavalla, Balamanohar Paluri
-
Publication number: 20190205372Abstract: In one embodiment, a method includes identifying a plurality of dyslexic users on an online social network. The plurality of dyslexic users may be identified based on content objects posted by these users over a particular time period, where the content objects may include one or more of word-level errors or sentence-level errors. A machine-learning model may be trained for text correction using a corpus of social network data, which may include at least the content objects with one or more of word-level errors or sentence-level errors, and a corresponding set of corrected content objects. A text string including one or more errors may be received from a client system associated with a first user. The text string may be transformed into a vector representation using an encoder of the machine-learning model. A corrected text string may be generated from the vector representation using a decoder of the machine-learning model.Type: ApplicationFiled: January 2, 2018Publication date: July 4, 2019Inventors: Xian Li, Irina-Elena Veliche, Debnil Sur, Shaomei Wu, Amit Bahl, Juan Miguel Pino
-
Patent number: 10157307Abstract: In one embodiment, a method includes capturing one or more images of a physical environment surrounding the first user; determining that one or more of the images comprises a face; generating a facial image by cropping a region of the image comprising the face; sending instructions to perform facial recognition on the facial image; receiving a name associated with a second user of the online social network, wherein the face is associated with the second user, wherein the first user is represented by a first node on a social graph maintained by the social-networking system and the second user is represented by a second node on the social graph, and the first node and the second node are connected by at least one edge; and providing an audible indication of the name associated with the second user.Type: GrantFiled: October 20, 2016Date of Patent: December 18, 2018Assignee: Facebook, Inc.Inventors: Paul Alexander Dow, Brett Alden Lavalla, Jeffrey C. Wieland, Shaomei Wu, Yuhang Zhao, Lindsay Reynolds, Matt King
-
Publication number: 20180181832Abstract: Systems, methods, and non-transitory computer-readable media can receive an image. One or more concepts depicted in the image are identified based on machine learning techniques. The one or more concepts are filtered based on filtering criteria to identify one or more selected concepts. An image description is generated comprising the one or more selected concepts.Type: ApplicationFiled: December 27, 2016Publication date: June 28, 2018Inventors: Shaomei Wu, Lada Ariana Adamic, Jeffrey C. Wieland, Omid Farivar, Hermes Germi Pique Corchs, Matt King, Brett Alden Lavalla, Balamanohar Paluri
-
Publication number: 20180114054Abstract: In one embodiment, a method includes capturing one or more images of a physical environment surrounding the first user; determining that one or more of the images comprises a face; generating a facial image by cropping a region of the image comprising the face; sending instructions to perform facial recognition on the facial image; receiving a name associated with a second user of the online social network, wherein the face is associated with the second user, wherein the first user is represented by a first node on a social graph maintained by the social-networking system and the second user is represented by a second node on the social graph, and the first node and the second node are connected by at least one edge; and providing an audible indication of the name associated with the second user.Type: ApplicationFiled: October 20, 2016Publication date: April 26, 2018Inventors: Paul Alexander Dow, Brett Alden Lavalla, Jeffrey C. Wieland, Shaomei Wu, Yuhang Zhao, Lindsay Reynolds, Matt King
-
Publication number: 20180032898Abstract: Systems, methods, and non-transitory computer-readable media can receive a plurality of comments to a posted content item. Each of the plurality of comments is associated with at least one category of a plurality of categories based on a machine learning model. A first comment of the plurality of comments is selected for inclusion in a comment sample to be presented in a graphical user interface based on the first comment being associated with a first category of the plurality of categories. A second comment of the plurality of comments is selected for inclusion in the comment sample based on the second comment being associated with a second category of the plurality of categories.Type: ApplicationFiled: July 27, 2016Publication date: February 1, 2018Inventors: Shaomei Wu, Isabel Kloumann, Lada Ariana Adamic, Erich James Owens
-
Publication number: 20170140056Abstract: Methods and systems for generating influence scores are disclosed. A plurality of opinions of a selected topic are retrieved from a database. Influencers each associated with at least one opinion from the plurality of opinions are determined using the processor. A score is automatically generated for each influencer using the processor, the score related to the selected topic.Type: ApplicationFiled: January 30, 2017Publication date: May 18, 2017Inventors: Chia-Wei Wu, Claire Cardie, Laurence Levy, Shaomei Wu, Vladimir Dmitrievich Barash, David Rusell Pierce
-
Patent number: 9558273Abstract: Methods and systems for generating influence scores are disclosed. A plurality of opinions of a selected topic are retrieved from a database. Influencers each associated with at least one opinion from the plurality of opinions are determined using the processor. A score is automatically generated for each influencer using the processor, the score related to the selected topic.Type: GrantFiled: September 23, 2013Date of Patent: January 31, 2017Assignee: Appinions Inc.Inventors: Chia-Wei Wu, Claire Cardie, Laurence Levy, Shaomei Wu, Vladimir Dmitrievich Barash, David Rusell Pierce
-
Publication number: 20140089323Abstract: Methods and systems for generating influence scores are disclosed. A plurality of opinions of a selected topic are retrieved from a database. Influencers each associated with at least one opinion from the plurality of opinions are determined using the processor. A score is automatically generated for each influencer using the processor, the score related to the selected topic.Type: ApplicationFiled: September 23, 2013Publication date: March 27, 2014Applicant: Appinions Inc.Inventors: Chia-Wei Wu, Claire Cardie, Laurence Levy, Shaomei Wu, Vladimir Dmitrievich Barash, David Rusell Pierce