Su Kung Chen, 69Saratoga, CA

6742161, May 25, 2004

Mar 7, 2000

09/520892

Barnaby James - Los Gatos CA
Su Chen - San Jose CA

ScanSoft, Inc. - Peabody MA

G06F 1721

715500, 715530

The present invention is a system and method for performing document recognition and processing in a distributed computing environment. The invention uses applications which are remotely located from one or more users and may be accessed via a network. One or more users utilize terminals including computers, facsimile machines, and/or scanners to transmit documents to be processed to a network or a network server which in turn transmits the documents to various computer software applications which process the documents at a network processing location. Once the documents have been processed, the processed documents are transmitted to the users according to one or more preferences associated with a user identification and/or authentication which may be determined by either a network server or an application server. Users utilizing a computer terminal make use of various data transfer programs capable of transferring document data over a network to an application server at a remote location and receiving processed document data via a network.

7558290, Jul 7, 2009

Dec 16, 2005

11/303651

Antonio Nucci - Burlingame CA, US
Su Chen - Somerset NJ, US

Narus, Inc. - Sunnyvale CA

H04J 3/18

370477

An important component of network monitoring is to collect traffic data which is a bottleneck due to large data size. We introduce a new table compression method called “Group Compression” to address this problem. This method uses a small training set to learn the relationship among columns and group them; the result is a “compression plan”. Based on this plan, each group is compressed separately. This method can reduce the compressed size to 60%-70% of the IP flow logs compressed by GZIP.

8516157, Aug 20, 2013

Apr 20, 2011

13/091090

Antonio Nucci - San Jose CA, US
Su Chen - Somerset NJ, US

Narus, Inc. - Sunnyvale CA

G06F 15/16
H04J 3/18

709247, 709246, 370477

The present invention relates to a method of compressing data in a network, the data comprising a plurality of packets each having a header and a payload, the header comprising a plurality of header fields, the method comprising generating a classification tree based on at least a portion of the plurality of header fields, determining a inter-packet compression plan based on the classification tree, and performing inter-packet compression in real time for each payload of at least a first portion of the plurality of packets, the inter-packet compression being performed according to at least a portion of the inter-packet compression plan.

8046496, Oct 25, 2011

Dec 12, 2007

11/955259

Antonio Nucci - San Jose CA, US
Su Chen - Sunnyvale CA, US

Narus, Inc. - Sunnyvale CA

G06F 15/16
H04J 3/18

709247, 709246, 370477

The present invention relates to a method of compressing data in a network, the data comprising a plurality of packets each having a header and a payload, the header comprising a plurality of header fields, the method comprising generating a classification tree based on at least a portion of the plurality of header fields, determining a inter-packet compression plan based on the classification tree, and performing inter-packet compression in real time for each payload of at least a first portion of the plurality of packets, the inter-packet compression being performed according to at least a portion of the inter-packet compression plan.

2022028, Sep 8, 2022

Feb 26, 2021

17/186436

- San Jose CA, US
Jianming Zhang - Campbell CA, US
Oliver Wang - Seattle WA, US
Simon Niklaus - San Jose CA, US
Mai Long - Portland OR, US
Su Chen - San Jose CA, US

G06T 7/593
G06T 7/30
G06T 7/13
G06T 7/143
G01S 17/894
G01S 17/42

This disclosure describes one or more implementations of a depth prediction system that generates accurate depth images from single input digital images. In one or more implementations, the depth prediction system enforces different sets of loss functions across mix-data sources to generate a multi-branch architecture depth prediction model. For instance, in one or more implementations, the depth prediction model utilizes different data sources having different granularities of ground truth depth data to robustly train a depth prediction model. Further, given the different ground truth depth data granularities from the different data sources, the depth prediction model enforces different combinations of loss functions including an image-level normalized regression loss function and/or a pair-wise normal loss among other loss functions.

2022027, Sep 1, 2022

Feb 26, 2021

17/186522

- San Jose CA, US
Jianming Zhang - Campbell CA, US
Oliver Wang - Seattle WA, US
Simon Niklaus - San Jose CA, US
Mai Long - Portland OR, US
Su Chen - San Jose CA, US

G06T 17/00
G06K 9/00
G06N 3/04
G06T 7/80

This disclosure describes implementations of a three-dimensional (3D) scene recovery system that reconstructs a 3D scene representation of a scene portrayed in a single digital image. For instance, the 3D scene recovery system trains and utilizes a 3D point cloud model to recover accurate intrinsic camera parameters from a depth map of the digital image. Additionally, the 3D point cloud model may include multiple neural networks that target specific intrinsic camera parameters. For example, the 3D point cloud model may include a depth 3D point cloud neural network that recovers the depth shift as well as include a focal length 3D point cloud neural network that recovers the camera focal length. Further, the 3D scene recovery system may utilize the recovered intrinsic camera parameters to transform the single digital image into an accurate and realistic 3D scene representation, such as a 3D point cloud.

2022019, Jun 23, 2022

Dec 18, 2020

17/126986

- San Jose CA, US
Su Chen - San Jose CA, US
Shuo Yang - Bloomington IN, US

G06T 7/11
G06T 7/143
G06T 7/136
G06T 7/162
G06T 7/90
G06N 3/08
G06N 3/04

The present disclosure relates to systems, non-transitory computer-readable media, and methods that utilize a deep neural network to process object user indicators and an initial object segmentation from a digital image to efficiently and flexibly generate accurate object segmentations. In particular, the disclosed systems can determine an initial object segmentation for the digital image (e.g., utilizing an object segmentation model or interactive selection processes). In addition, the disclosed systems can identify an object user indicator for correcting the initial object segmentation and generate a distance map reflecting distances between pixels of the digital image and the object user indicator. The disclosed systems can generate an image-interaction-segmentation triplet by combining the digital image, the initial object segmentation, and the distance map. By processing the image-interaction-segmentation triplet utilizing the segmentation neural network, the disclosed systems can provide an updated object segmentation for display to a client device.

2023012, Apr 27, 2023

Jan 31, 2022

17/589114

- San Jose CA, US
Su Chen - San Jose CA, US
Scott Cohen - Sunnyvale CA, US
Zhe Lin - Fremont CA, US
Zijun Wei - San Jose CA, US
Jianming Zhang - Campbell CA, US

G06V 10/82
G06N 3/08
G06T 7/00

The present disclosure relates to systems, methods, and non-transitory computer-readable media that generates object masks for digital objects portrayed in digital images utilizing a detection-masking neural network pipeline. In particular, in one or more embodiments, the disclosed systems utilize detection heads of a neural network to detect digital objects portrayed within a digital image. In some cases, each detection head is associated with one or more digital object classes that are not associated with the other detection heads. Further, in some cases, the detection heads implement multi-scale synchronized batch normalization to normalize feature maps across various feature levels. The disclosed systems further utilize a masking head of the neural network to generate one or more object masks for the detected digital objects. In some cases, the disclosed systems utilize post-processing techniques to filter out low-quality masks.