Invention Grant
- Patent Title: Neural network architecture system for deep odometry assisted by static scene optical flow
-
Application No.: US15703879Application Date: 2017-09-13
-
Publication No.: US10671083B2Publication Date: 2020-06-02
- Inventor: Wentao Zhu , Yi Wang , Yi Luo
- Applicant: TuSimple, Inc.
- Applicant Address: US CA San Diego
- Assignee: TUSIMPLE, INC.
- Current Assignee: TUSIMPLE, INC.
- Current Assignee Address: US CA San Diego
- Agent Paul Liu; Vinay Sathe
- Main IPC: G05D1/02
- IPC: G05D1/02 ; G06K9/46 ; G01C21/32 ; G06K9/62 ; G06T7/246 ; G01C21/16 ; G01S17/89 ; G06N5/02 ; G06N20/00 ; G06T7/269 ; G01C22/00 ; G01S17/42 ; G01S7/48 ; G01S17/86

Abstract:
A system for visual odometry is disclosed. The system includes: an internet server, comprising: an I/O port, configured to transmit and receive electrical signals to and from a client device; a memory; one or more processing units; and one or more programs stored in the memory and configured for execution by the one or more processing units, the one or more programs including instructions for: extracting representative features from a pair input images in a first convolution neural network (CNN) in a visual odometry model; merging, in a first merge module, outputs from the first CNN; decreasing feature map size in a second CNN; generating a first flow output for each layer in a first deconvolution neural network (DNN); merging, in a second merge module, outputs from the second CNN and the first DNN; generating a second flow output for each layer in a second DNN; and reducing accumulated errors in a recurrent neural network (RNN).
Public/Granted literature
- US20190079534A1 NEURAL NETWORK ARCHITECTURE SYSTEM FOR DEEP ODOMETRY ASSISTED BY STATIC SCENE OPTICAL FLOW Public/Granted day:2019-03-14
Information query