# Transformer-OCR

**Repository Path**: chenyang918/Transformer-OCR

## Basic Information

- **Project Name**: Transformer-OCR
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-09-06
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

Scene Text Recognition via Transformer


model architecture:

<img src="art.png" width = 60%  div align=center />

Results across a number of methods and datasets:

<img src="result.png" width = 60%  div align=center />
Heat map of the source attention (encoder memory) score of the first layer of decoder:

<img src="heatmap.png" width = 70%  div align=center />

The transformer source code from:http://nlp.seas.harvard.edu/2018/04/03/attention.html