Skip to content

Releases: breezedeus/CnSTD

feat: PP-OCRv6 multilingual support

Choose a tag to compare

@breezedeus breezedeus released this 05 Jul 01:20
e8d5d0c

Update 2026.07.04:发布 V1.2.8

Major Changes:

  • Added RapidOCR-based PP-OCRv6 text detection models:
    multi_PP-OCRv6_det_tiny, multi_PP-OCRv6_det_small, and
    multi_PP-OCRv6_det_medium.
  • Changed the default detection model to multi_PP-OCRv6_det_small.
  • Added PP-OCRv6 language selection through lang_type; multi is the model
    family name and is not a valid lang_type.
  • Added HuggingFace model repositories for PP-OCRv6 detection models.
  • Standardized CnSTD logging for library and CLI usage.

主要变更:

  • 基于 RapidOCR 支持 PP-OCRv6 文本检测模型:
    multi_PP-OCRv6_det_tinymulti_PP-OCRv6_det_small
    multi_PP-OCRv6_det_medium
  • 默认检测模型改为 multi_PP-OCRv6_det_small
  • 支持通过 lang_type 指定 PP-OCRv6 语言类型;multi 是模型族名称,不是可传入的
    lang_type
  • 新增 PP-OCRv6 检测模型对应的 HuggingFace 模型仓库。
  • 规范 CnSTD 在库调用和 CLI 场景下的日志行为。

Improve robustness of subprocess calls in YOLOv7 modules

Choose a tag to compare

@breezedeus breezedeus released this 01 May 02:00
e5f1946

Update 2026.05.01:发布 V1.2.7.3

Bug Fixes:

  • Improve robustness of subprocess calls in YOLOv7 modules by avoiding shell-based invocations in git and pip commands

Bug Fixes:

  • 提升 YOLOv7 模块中子进程调用的稳健性,避免在 git 和 pip 命令中依赖 shell 调用

fix: support rapidocr model_root_dir

Choose a tag to compare

@breezedeus breezedeus released this 28 Apr 13:33
1c0ff11

Update 2026.04.28:发布 V1.2.7.2

Bug Fixes:

  • Support model_root_dir in RapidDetector so RapidOCR model assets can be resolved from a custom model root directory

Bug Fixes:

  • 修复 RapidDetectormodel_root_dir 的支持,确保 RapidOCR 相关模型资源可从自定义模型根目录正确解析

Bugfixes

Choose a tag to compare

@breezedeus breezedeus released this 07 Feb 09:11
8196f44

Update 2026.02.07:发布 V1.2.7.1

Bug Fixes:

  • Align set_logger with the module logger so configured handlers/levels apply to downloads

Bug Fixes:

  • 让 set_logger 使用模块级 logger,确保下载日志遵循配置的 handler 和 level

Update 2026.02.07:发布 V1.2.7

Bug Fixes:

  • Switch model download to use HuggingFaceDownloader instead of shelling out to huggingface-cli

Bug Fixes:

  • 模型下载逻辑改为使用 HuggingFaceDownloader,避免调用 huggingface-cli

Bug Fixed

Choose a tag to compare

@breezedeus breezedeus released this 27 Jun 00:02
bf623a2

Update 2025.06.27:发布 V1.2.6.1

Major Changes:

  • Bug Fixed

主要变更:

  • 修复已知 bug

Integrated the latest PPOCRv5 text detection models

Choose a tag to compare

@breezedeus breezedeus released this 26 Jun 14:49
943f04d

Update 2025.06.25:发布 V1.2.6

Major Changes:

  • Integrated the latest PPOCRv5 text detection functionality based on RapidOCR for even faster inference speed
    • Added support for PP-OCRv5 detection models: ch_PP-OCRv5_det and ch_PP-OCRv5_det_server
  • Fixed some known bugs

主要变更:

  • 基于 RapidOCR 集成 PPOCRv5 最新版文本检测功能,提供更快的推理速度
    • 新增支持 PP-OCRv5 检测模型:ch_PP-OCRv5_detch_PP-OCRv5_det_server
  • 修复部分已知 bug

Bug Fixes

Choose a tag to compare

@breezedeus breezedeus released this 08 Dec 08:56
47eae63

Update 2024.12.08:发布 V1.2.5.2

Bug Fixes:

  • Fix compatibility issue of setting environment variables on Windows systems
  • Use subprocess.run instead of os.system for better cross-platform support

Bug Fixes:

  • 修复在 Windows 系统下设置环境变量的兼容性问题
  • 使用 subprocess.run 替代 os.system 以提供更好的跨平台支持

Update 2024.11.30:发布 V1.2.5.1

Choose a tag to compare

@breezedeus breezedeus released this 30 Nov 01:55
6811470

Update 2024.11.30:发布 V1.2.5.1

Major Changes:

  • en_PP-OCRv3_det still uses the previous version and does not use RapidDetector

Bug Fixes:

  • en_PP-OCRv3_det 依旧使用之前的版本,不使用 RapidDetector

v1.2.5: Merge pull request #88 from breezedeus/pytorch

Choose a tag to compare

@breezedeus breezedeus released this 24 Nov 13:07
f4dd500

Update 2024.11.24:发布 V1.2.5

Major Changes:

  • Integrated latest PPOCRv4 text detection functionality based on RapidOCR for faster inference
    • Added support for PP-OCRv4 detection models, including standard and server versions
    • Added support for PP-OCRv3 English detection model
  • Optimized model download functionality with support for domestic mirrors

主要变更:

  • 基于 RapidOCR 集成 PPOCRv4 最新版文本检测功能,提供更快的推理速度
    • 新增支持 PP-OCRv4 检测模型,包括标准版和服务器版
    • 新增支持 PP-OCRv3 英文检测模型
  • 优化模型下载功能,支持从国内镜像下载模型文件

新模型名称:

model_name PyTorch 版本 ONNX 版本 支持检测的语言 模型文件大小
en_PP-OCRv3_det X 英文、数字 2.3 M
ch_PP-OCRv4_det X 简体中问、英文、数字 4.5 M
ch_PP-OCRv4_det_server X 简体中问、英文、数字 108 M

使用方式

  • 命令行:$ cnstd predict -m ch_PP-OCRv4_det_server -i examples/1_res.png -o outputs
  • Python:
    from cnstd import CnStd
    
    std = CnStd(model_name="ch_PP-OCRv4_det_server")
    out = std.detect("examples/1_res.png")
    print(out)

Fix: some formats of models require fixed-size input images

Choose a tag to compare

@breezedeus breezedeus released this 22 Jun 15:30
7de2abb

Update 2024.06.22:发布 V1.2.4.2

Major Changes:

  • Added a new parameter static_resized_shape when initializing YoloDetector, which is used to resize the input image to a fixed size. Some formats of models require fixed-size input images during inference, such as CoreML.

主要变更:

  • YoloDetector 初始化时加入了参数 static_resized_shape, 用于把输入图片 resize 为固定大小。某些格式的模型在推理时需要固定大小的输入图片,如 CoreML