2 Pull requests merged by 1 person

• update logo.jpg

  #16204 merged Aug 7, 2025

• replace logo for docs

  #16201 merged Aug 7, 2025

3 Pull requests opened by 3 people

• Bump actions/checkout from 4 to 5 in the github-actions group

  #16219 opened Aug 12, 2025

• support multi-language for PP-StructureV3

  #16228 opened Aug 13, 2025

• add en, th, el models

  #16229 opened Aug 13, 2025

11 Issues closed by 5 people

• save gpu log failed

  #16191 closed Aug 12, 2025

• 飞腾 2000+ 服务器跑 v5 模型报错，报段错误 请大佬帮我看看，啥问题，感谢了

  #15843 closed Aug 12, 2025

• ValueError in PP_structureV3

  #15967 closed Aug 11, 2025

• 我用VS2019编译ppocr时遇到YAML无法解析的问题

  #15562 closed Aug 9, 2025

• ocr使用gpu识别的时候,不时会有超长耗时出现

  #15166 closed Aug 9, 2025

• Paddleocr3.0.0版本运行报错

  #15422 closed Aug 9, 2025

• PaddleOCR == 3.0.0 / Enable word-level detection (not line-level)

  #15760 closed Aug 9, 2025

• 算法、图像增强问题

  #15978 closed Aug 9, 2025

• [Regression] Lack of Slice Parameter

  #15868 closed Aug 8, 2025

• 指定输出结果集

  #15973 closed Aug 8, 2025

• 显卡支持

  #16198 closed Aug 7, 2025

8 Issues opened by 8 people

• 运行onnx模型推理时，报错"invalid input 'img' in NormalizeImage"

  #16232 opened Aug 14, 2025

• Spaces between words/numbers missing in pp_ocr_v5 (v3.1.0)

  #16231 opened Aug 13, 2025

• Load model (ABINet)

  #16227 opened Aug 13, 2025

• Android Demo各种编译不过，能否提供一个符合新版本的gradle版本的可以编译通过的demo

  #16211 opened Aug 8, 2025

• 同样的微调数据集，v5效果比v4差

  #16210 opened Aug 8, 2025

• c++调用paddleocr v3.1.0（paddlepaddle v3.1.0 ），内存一直上涨

  #16205 opened Aug 7, 2025

• PaddleOCR不支持RTX50系列

  #16200 opened Aug 7, 2025

• PaddleOCR报错PreconditionNotMetError: Tensor holds no memory. Call Tensor::mutable_data firstly

  #16199 opened Aug 7, 2025

19 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• det模型训练的时候eval会爆显存

  #16195 commented on Aug 7, 2025 • 0 new comments

• paddleocr无法识别图片中的文字

  #16160 commented on Aug 7, 2025 • 0 new comments

• 识别有较多空白的图片有问题

  #16062 commented on Aug 8, 2025 • 0 new comments

• 3.0版本训练时无法创建train.log

  #15931 commented on Aug 8, 2025 • 0 new comments

• v5识别模型当batch=1的时候，测试的图片很长，等比缩放之后，总是导致显存不够，能否限制图片宽度，然后再去缩放？而不是直接resize

  #15848 commented on Aug 8, 2025 • 0 new comments

• PaddleOCR在华为晟腾芯片上的精度问题

  #16159 commented on Aug 8, 2025 • 0 new comments

• Export model Error: 'NRTRLabelDecode'

  #15901 commented on Aug 8, 2025 • 0 new comments

• RuntimeError: std::exception

  #16196 commented on Aug 8, 2025 • 0 new comments

• 运行的时候会一直卡在 Fetching 6 files 100% 的进度

  #16193 commented on Aug 9, 2025 • 0 new comments

• 5060Ti train error

  #15720 commented on Aug 9, 2025 • 0 new comments

• official cmd sample failed on mac

  #16179 commented on Aug 10, 2025 • 0 new comments

• ppocrv5 推理显存持续升高

  #16011 commented on Aug 11, 2025 • 0 new comments

• paddlex --serve --pipeline OCR 的坐标跟之前 区别很大

  #16137 commented on Aug 11, 2025 • 0 new comments

• DocImgOrientationClassification模块模型路径不支持中文

  #16183 commented on Aug 11, 2025 • 0 new comments

• 使用paddle-ocr V5的3.0.0版本,微调模型时，卡死在[2025/06/28 17:34:57] ppocr INFO: During the training process, after the 0th iteration, an evaluation is run every 1500 iterations

  #15894 commented on Aug 11, 2025 • 0 new comments

• the dataset load, sometime run into dead loops

  #14986 commented on Aug 13, 2025 • 0 new comments

• DecodeImage中可能存在的错误

  #14698 commented on Aug 13, 2025 • 0 new comments

• Paddleocr3.0.0 版本运行报错

  #15558 commented on Aug 13, 2025 • 0 new comments

• faq: how to change font

  #16129 commented on Aug 8, 2025 • 0 new comments