Merge pull request #7102 from MissPenguin/release/2.5

refine doc
pull/7134/head
MissPenguin 2022-08-05 14:52:37 +08:00 committed by GitHub
commit 3a41010fa3
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 4 additions and 4 deletions

View File

@ -22,7 +22,7 @@ PaddleOCR场景应用覆盖通用制造、金融、交通行业的主要OCR
| 类别 | 亮点 | 模型下载 | 教程 | | 类别 | 亮点 | 模型下载 | 教程 |
| ---------------------- | ------------ | -------------- | --------------------------------------- | | ---------------------- | ------------ | -------------- | --------------------------------------- |
| 高精度中文识别模型SVTR | 新增模型 | [模型下载](#2) | [中文](./高精度中文识别模型.md)/English | | 高精度中文识别模型SVTR | 比PP-OCRv3识别模型精度高3%,可用于数据挖掘或对预测效率要求不高的场景。| [模型下载](#2) | [中文](./高精度中文识别模型.md)/English |
| 手写体识别 | 新增字形支持 | | | | 手写体识别 | 新增字形支持 | | |
<a name="12"></a> <a name="12"></a>

View File

@ -2,7 +2,7 @@
## 1. 简介 ## 1. 简介
PP-OCRv3是百度开源的超轻量级场景文本检测识别模型库其中超轻量的场景中文识别模型SVTR_LCNet使用了SVTR算法结构。为了保证速度SVTR_LCNet将SVTR模型的Local Blocks替换为LCNet使用两层Global Blocks。在中文场景中PP-OCRv3识别主要使用如下优化策略 PP-OCRv3是百度开源的超轻量级场景文本检测识别模型库其中超轻量的场景中文识别模型SVTR_LCNet使用了SVTR算法结构。为了保证速度SVTR_LCNet将SVTR模型的Local Blocks替换为LCNet使用两层Global Blocks。在中文场景中PP-OCRv3识别主要使用如下优化策略[详细技术报告](../doc/doc_ch/PP-OCRv3_introduction.md)
- GTCAttention指导CTC训练策略 - GTCAttention指导CTC训练策略
- TextConAug挖掘文字上下文信息的数据增广策略 - TextConAug挖掘文字上下文信息的数据增广策略
- TextRotNet自监督的预训练模型 - TextRotNet自监督的预训练模型

View File

@ -185,7 +185,7 @@ UDMLUnified-Deep Mutual Learning联合互学习是PP-OCRv2中就采用的
**6UIM无标注数据挖掘方案** **6UIM无标注数据挖掘方案**
UIMUnlabeled Images Mining是一种非常简单的无标注数据挖掘方案。核心思想是利用高精度的文本识别大模型对无标注数据进行预测获取伪标签并且选择预测置信度高的样本作为训练数据用于训练小模型。使用该策略识别模型的准确率进一步提升到79.4%+1%)。 UIMUnlabeled Images Mining是一种非常简单的无标注数据挖掘方案。核心思想是利用高精度的文本识别大模型对无标注数据进行预测获取伪标签并且选择预测置信度高的样本作为训练数据用于训练小模型。使用该策略识别模型的准确率进一步提升到79.4%+1%)。实际操作中我们使用全量数据集训练高精度SVTR-Tiny模型acc=82.5%)进行数据挖掘,点击获取[模型下载地址和使用教程](../../applications/高精度中文识别模型.md)。
<div align="center"> <div align="center">
<img src="../ppocr_v3/UIM.png" width="500"> <img src="../ppocr_v3/UIM.png" width="500">

View File

@ -200,7 +200,7 @@ UDML (Unified-Deep Mutual Learning) is a strategy proposed in PP-OCRv2 which is
**6UIMUnlabeled Images Mining** **6UIMUnlabeled Images Mining**
UIM (Unlabeled Images Mining) is a very simple unlabeled data mining strategy. The main idea is to use a high-precision text recognition model to predict unlabeled images to obtain pseudo-labels, and select samples with high prediction confidence as training data for training lightweight models. Using this strategy, the accuracy of the recognition model is further improved to 79.4% (+1%). UIM (Unlabeled Images Mining) is a very simple unlabeled data mining strategy. The main idea is to use a high-precision text recognition model to predict unlabeled images to obtain pseudo-labels, and select samples with high prediction confidence as training data for training lightweight models. Using this strategy, the accuracy of the recognition model is further improved to 79.4% (+1%). In practice, we use the full data set to train the high-precision SVTR_Tiny model (acc=82.5%) for data mining. [SVTR_Tiny model download and tutorial](../../applications/高精度中文识别模型.md).
<div align="center"> <div align="center">
<img src="../ppocr_v3/UIM.png" width="500"> <img src="../ppocr_v3/UIM.png" width="500">