Patents by Inventor Cuiling Lan
Cuiling Lan 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: 20230388548Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: ApplicationFiled: August 9, 2023Publication date: November 30, 2023Applicant: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Patent number: 11765390Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: GrantFiled: August 12, 2022Date of Patent: September 19, 2023Assignee: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Publication number: 20220385944Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: ApplicationFiled: August 12, 2022Publication date: December 1, 2022Applicant: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Patent number: 11451827Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: GrantFiled: April 14, 2021Date of Patent: September 20, 2022Assignee: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Publication number: 20210235122Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: ApplicationFiled: April 14, 2021Publication date: July 29, 2021Applicant: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Patent number: 11003949Abstract: Various implementations of the subject matter described herein relate to a neural network-based action detection. There is provided an action detection scheme using a neural network. The action detection scheme can design and optimize the neural network model based on respective importance of different frames such that frames that are more important or discriminative for action recognition tend to be assigned with higher weights and frames that are less important or discriminative for action recognition tend to be assigned with lower weights.Type: GrantFiled: October 31, 2017Date of Patent: May 11, 2021Assignee: MICROSOFT TECHNOLOGY LICENSING, LLCInventors: Cuiling Lan, Wenjun Zeng, Sijie Song, Junliang Xing
-
Patent number: 11006149Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: GrantFiled: March 17, 2020Date of Patent: May 11, 2021Assignee: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Patent number: 10869029Abstract: In accordance with implementations of the subject matter described herein, a hybrid digital-analog coding scheme is proposed. In general, in accordance with implementations of the subject matter described herein, digital or analog encoding is selected at a level of chunks of a frame rather than at a level of the whole frame. Further, the encoding of a chunk is based on an expected distortion to be caused by analog transmission of the chunk over a communication channel, the distortion being estimated based on a constraint on available transmission resources over the transmission channel.Type: GrantFiled: July 20, 2017Date of Patent: December 15, 2020Assignee: Microsoft Technology Licensing, LLCInventors: Cuiling Lan, Chong Luo, Feng Wu, Wenjun Zeng
-
Patent number: 10789482Abstract: In implementations of the subject matter described herein, an action detection scheme using a recurrent neural network (RNN) is proposed. Representation information of an incoming frame of a video and a predefined action label for the frame are obtained to train a learning network including RNN elements and a classification element. The representation information represents an observed entity in the frame. Specifically, parameters for the RNN elements are determined based on the representation information and the predefined action label. With the determined parameters, the RNN elements are caused to extract features for the frame based on the representation information and features for a preceding frame. Parameters for the classification element are determined based on the extracted features and the predefined action label. The classification element with the determined parameters generates a probability of the frame being associated with the predefined action label.Type: GrantFiled: March 28, 2017Date of Patent: September 29, 2020Assignee: Microsoft Technology Licensing, LLCInventors: Cuiling Lan, Wenjun Zeng, Yanghao Li, Junliang Xing
-
Publication number: 20200221129Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: ApplicationFiled: March 17, 2020Publication date: July 9, 2020Applicant: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Patent number: 10623776Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: GrantFiled: April 9, 2019Date of Patent: April 14, 2020Assignee: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Publication number: 20200074227Abstract: Various implementations of the subject matter described herein relate to a neural network-based action detection. There is provided an action detection scheme using a neural network. The action detection scheme can design and optimize the neural network model based on respective importance of different frames such that frames that are more important or discriminative for action recognition tend to be assigned with higher weights and frames that are less important or discriminative for action recognition tend to be assigned with lower weights.Type: ApplicationFiled: October 31, 2017Publication date: March 5, 2020Applicant: Microsoft Technology Licensing, LLCInventors: Cuiling LAN, Wenjun ZENG, Sijie SONG, Junliang XING
-
Patent number: 10460456Abstract: In implementations of the subject matter described herein, a current captured frame of a video is compared with a respective reference frame to determine a correlation therebetween. The correlation is used to indicate a change degree of the current frame. If the correlation for the current frame is below a predetermined threshold, the current frame may not be directly determined as including a motion of an object. Instead, correlations between one or more frames before or after the current frame and their respective reference frames are took into account. If the correlations of the frames under consideration are below the predetermined threshold, it may be detected that the current frame includes a motion of an object. In this way, incorrect detection of the object motion is reduced in the cases when larger changes in frames of a video are caused by factors such as noise and error, and the accuracy of the object motion detection is improved.Type: GrantFiled: November 2, 2016Date of Patent: October 29, 2019Assignee: Microsoft Technology Licensing, LLCInventors: Wenxuan Xie, Cuiling Lan, Wenjun Zeng
-
Publication number: 20190306531Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: ApplicationFiled: April 9, 2019Publication date: October 3, 2019Applicant: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Patent number: 10298955Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: GrantFiled: December 12, 2017Date of Patent: May 21, 2019Assignee: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Publication number: 20190141316Abstract: In accordance with implementations of the subject matter described herein, a hybrid digital-analog coding scheme is proposed. In general, in accordance with implementations of the subject matter described herein, digital or analog encoding is selected at a level of chunks of a frame rather than at a level of the whole frame. Further, the encoding of a chunk is based on an expected distortion to be caused by analog transmission of the chunk over a communication channel, the distortion being estimated based on a constraint on available transmission resources over the transmission channel.Type: ApplicationFiled: July 20, 2017Publication date: May 9, 2019Inventors: Cuiling Lan, Chong Luo, Feng Wu, Wenjun Zeng
-
Publication number: 20190080176Abstract: In implementations of the subject matter described herein, an action detection scheme using a recurrent neural network (RNN) is proposed. Representation information of an incoming frame of a video and a predefined action label for the frame are obtained to train a learning network including RNN elements and a classification element. The representation information represents an observed entity in the frame. Specifically, parameters for the RNN elements are determined based on the representation information and the predefined action label. With the determined parameters, the RNN elements are caused to extract features for the frame based on the representation information and features for a preceding frame. Parameters for the classification element are determined based on the extracted features and the predefined action label. The classification element with the determined parameters generates a probability of the frame being associated with the predefined action label.Type: ApplicationFiled: March 28, 2017Publication date: March 14, 2019Inventors: Cuiling LAN, Wenjun ZENG, Yanghao LI, Junliang XING
-
Patent number: 10019629Abstract: In implementations of the subject matter described herein, an action detection scheme using a recurrent neural network (RNN) is proposed. Joint locations for a skeleton representation of an observed entity in a frame of a video and a predefined action label for the frame are obtained to train a learning network including RNN elements and a classification element. Specifically, first weights for mapping the joint locations to a first feature for the frame generated by a first RNN element in a learning network and second weights for mapping the joint locations to a second feature for the frame generated by a second RNN element in the learning network are determined based on the joint locations and the predefined action label. The first and second weights are determined by increasing a first correlation between the first feature and a first subset of the joint locations and a second correlation between the second feature and the first subset of the joint locations.Type: GrantFiled: May 31, 2016Date of Patent: July 10, 2018Assignee: Microsoft Technology Licensing, LLCInventors: Cuiling Lan, Wenjun Zeng, Wentao Zhu, Junliang Xing
-
Publication number: 20180103270Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: ApplicationFiled: December 12, 2017Publication date: April 12, 2018Applicant: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan
-
Patent number: 9866868Abstract: Techniques for selectively transforming one or more coding units when coding video content are described herein. The techniques may include determining whether or not to transform a particular coding unit. The determination may be based on a difference in pixel values of the particular coding unit and/or one or more predefined rate-distortion constraints. When it is determined to not perform a transform, the particular coding unit may be coded without transforming the particular coding unit.Type: GrantFiled: June 16, 2017Date of Patent: January 9, 2018Assignee: Microsoft Technology Licensing, LLCInventors: Jizheng Xu, Cuiling Lan