Chen Kuo Sun

US Patent:

20220406001, Dec 22, 2022

Filed:

Jun 21, 2021

Appl. No.:

17/353296

Inventors:

- Grand Cayman, KY
Haiying Cheng - Culver City CA, US
Chen Sun - Los Angeles CA, US

International Classification:

G06T 15/04
G06K 9/00
G06T 17/20
G06T 13/20
G10L 25/48
G10L 25/03

Abstract:

Systems and methods for generating a virtual article of clothing at a display are described. Some examples may include: obtaining video data and audio data, analyzing the video data to determine one or more body joints of a target object appearing in the video data. A mesh based on the determined one or more body joints may be generated. The audio data may be analyzed to determine audio characteristics associated with the audio data. Texture rendering information associated with a virtual article of clothing may be determined based on the audio characteristics. A rendered video may be generated by rendering the virtual article of clothing to the generated mesh using the texture rendering information.

US Patent:

20220406337, Dec 22, 2022

Filed:

Jun 21, 2021

Appl. No.:

17/353441

Inventors:

- Grand Cayman, KY
Zihao Chen - Los Angeles CA, US
Chen Sun - Los Angeles CA, US

International Classification:

G11B 27/031

Abstract:

Systems and methods for rendering a segmentation contour effect are described. More specifically, video data including one or more video frames and audio data are obtained. Based on the video data, one or more segments in each of the one or more video frame are determined. The audio data is analyzed to determine beat characteristics of each beat. A segmentation contour effect to be applied to the one or more segments in the video data is determined based on the beat characteristics. A rendered video is generated by synchronizing the segmentation contour effect to audio data.

US Patent:

20220224433, Jul 14, 2022

Filed:

Jan 24, 2022

Appl. No.:

17/583185

Inventors:

- Santa Clara CA, US
Alexandra Wright - Oakland CA, US
Chen Sun - Berkeley CA, US
Mark Wade - Berkeley CA, US
Roy Edward Meade - Lafayette CA, US

International Classification:

H04J 14/02
H04Q 11/00
H04B 10/50

Abstract:

A TORminator module is disposed with a switch linecard of a rack. The TORminator module receives downlink electrical data signals from a rack switch. The TORminator module translates the downlink electrical data signals into downlink optical data signals. The TORminator module transmits multiple subsets of the downlink optical data signals through optical fibers to respective SmartDistributor modules disposed in respective racks. Each SmartDistributor module receives multiple downlink optical data signals through a single optical fiber from the TORminator module. The SmartDistributor module demultiplexes the multiple downlink optical data signals and distributes them to respective servers. The SmartDistributor module receives multiple uplink optical data signals from multiple servers and multiplexes them onto a single optical fiber for transmission to the TORminator module. The TORminator module coverts the multiple uplink optical data signals to multiple uplink electrical data signals, and transmits the multiple uplink electrical data signals to the rack switch.

US Patent:

20210264203, Aug 26, 2021

Filed:

Nov 18, 2019

Appl. No.:

17/046313

Inventors:

- Mountain View CA, US
Aleksei Timofeev - Mountain View CA, US
Zhen Li - Sunnyvale CA, US
Chun-Ta Lu - Sunnyvale CA, US
Manan Shah - Los Altos CA, US
Chen Sun - San Francisco CA, US
Krishnamurthy Viswanathan - Sunnyvale CA, US
Chao Jia - Sunnyvale CA, US

International Classification:

G06K 9/62
G06K 9/46

Abstract:

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for realizing a multimodal image classifier. In an aspect, a method includes, for each image of a plurality of images: processing the image by a textual generator model to obtain a set of phrases that are descriptive of the content of the image, wherein each phrase is one or more terms, processing the set of phrases by a textual embedding model to obtain an embedding of predicted text for the image, and processing the image using an image embedding model to obtain an embedding of image pixels of the image. Then a multimodal image classifier is trained on the embeddings of predicted text for the images and the embeddings of image pixels for the images to produce, as output, labels of an output taxonomy to classify an image based on the image as input.

US Patent:

20210150350, May 20, 2021

Filed:

Nov 16, 2020

Appl. No.:

17/099656

Inventors:

- Mountain View CA, US
Yi Shen - Sunnyvale CA, US
Hang Zhao - Sunnyvale CA, US
Chen Sun - San Francisco CA, US

International Classification:

G06N 3/08

Abstract:

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for agent trajectory prediction using vectorized inputs.

US Patent:

20200310035, Oct 1, 2020

Filed:

Mar 26, 2020

Appl. No.:

16/831734

Inventors:

- Emeryville CA, US
John Fini - Oakland CA, US
Derek Van Orden - San Francisco CA, US
Chen Sun - Berkeley CA, US
Mark Wade - Berkeley CA, US

International Classification:

G02B 6/293
G02F 1/225

Abstract:

A ring resonator device includes a passive optical cavity having a circuitous configuration into which is built a photodetector device. The photodetector device includes a first implant region formed within the passive optical cavity that includes a first type of implanted doping material. The photodetector device includes a second implant region formed within the passive optical cavity that includes a second type of implanted doping material, where the second type of implanted doping material is different than the first type of implanted doping material. The photodetector device includes an intrinsic absorption region present within the passive optical cavity between the first implant region and the second implant region. A first electrical contact is electrically connected to the first implant region and to a detecting circuit. A second electrical contact is electrically connected to the second implant region and to the detecting circuit.

US Patent:

20200125112, Apr 23, 2020

Filed:

Oct 22, 2018

Appl. No.:

16/167007

Inventors:

- Mountain View CA, US
Congcong Li - Cupertino CA, US
Alper Ayvaci - Santa Clara CA, US
Chen Sun - San Francisco CA, US
Kevin Murphy - Palo Alto CA, US
Ruichi Yu - College Park MD, US

International Classification:

G05D 1/02
G05D 1/00
G01S 17/93
B60W 30/095
G06K 9/00
G06K 9/62

Abstract:

Aspects of the disclosure relate to training and using a model for identifying actions of objects. For instance, LIDAR sensor data frames including an object bounding box corresponding to an object as well as an action label for the bounding box may be received. Each sensor frame is associated with a timestamp and is sequenced with respect to other sensor frames. Each given sensor data frame may be projected into a camera image of the object based on the timestamp associated with the given sensor data frame in order to provide fused data. The model may be trained using the fused data such that the model is configured to, in response to receiving fused data, the model outputs an action label for each object bounding box of the fused data. This output may then be used to control a vehicle in an autonomous driving mode.

US Patent:

20170046613, Feb 16, 2017

Filed:

Apr 5, 2016

Appl. No.:

15/091490

Inventors:

- Menlo Park CA, US
Lubomir Bourdev - Mountain View CA, US
Ronan Stéfan Collobert - Los Altos CA, US
Chen Sun - Los Angeles CA, US

International Classification:

G06N 3/04
G06N 3/08

Abstract:

Systems, methods, and non-transitory computer-readable media can obtain a content item to be evaluated by a set of cascaded convolutional neural networks, the set of cascaded convolutional neural networks including at least a first convolutional neural network (CNN) and a second CNN. The content item can be provided to the first CNN as input, wherein an output of the first CNN includes data describing at least one region of interest in the content item and at least one first concept corresponding to the region of interest. The output of the first CNN can be provided to the second CNN as input, wherein an output of the second CNN includes data describing at least one second concept corresponding to the region of interest, the second concept being more accurate than the first concept.