Publication Type
Patent
Version
publishedVersion
Publication Date
7-2022
Abstract
A system uses a neural network based model to perform scene text recognition. The system achieves high accuracy of prediction of text from scenes based on a neural network architecture that uses double attention mechanism. The neural network based model includes a convolutional neural network component that outputs a set of visual features and an attention extractor neural network component that determines attention scores based on the visual features. The visual features and the attention scores are combined to generate mixed features that are provided as input to a character recognizer component that determines a second attention score and recognizes the characters based on the second attention score. The system trains the neural network based model by adjusting the neural network parameters to minimize a multi-class gradient harmonizing mechanism (GHM) loss. The multi-class GHM loss varies based on a level of difficulty of the sample.
Discipline
Software Engineering | Systems Architecture
Research Areas
Software and Cyber-Physical Systems
First Page
1
Last Page
18
Publisher
US Patent Office
Embargo Period
12-17-2024
Citation
ZHOU, Pan; TANG, Peng; XU, Ran; and HOI, Steven Chu Hong.
Neural network based scene text recognition [US Patent US 2022/0237403 A1]. (2022). 1-18.
Available at: https://ink.library.smu.edu.sg/sis_research/9808
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Additional URL
https://patents.google.com/patent/US20220237403A1