PP-OCR Industry Landing: from Training to Deployment.PP-OCR Series Model List（Update on September 8th） Model introductionĬhinese and English ultra-lightweight PP-OCRv2 model（11.6M）Ĭhinese and English ultra-lightweight PP-OCR model (9.4M)Ĭhinese and English general PP-OCR model (143.4M)įor more model downloads (including multiple languages), please refer to PP-OCR series model downloads.įor a new language request, please refer to Guideline for new language_requests. Support Linux, Windows, MacOS and other systems.Support user-defined training, provides rich predictive inference deployment solutions.Data synthesis tool, i.e., Style-Text: easy to synthesize a large number of images which are similar to the target scene image.
Semi-automatic data annotation tool, i.e., PPOCRLabel: support fast and efficient data annotation.Support multi-language recognition: Korean, Japanese, German, French.Support Chinese, English, and digit recognition, vertical text recognition, and long text recognition.
General PP-OCR server series models: detection (47.1M) + direction classifier (1.4M) + recognition (94.9M) = 143.4M.Ultra lightweight PP-OCR mobile series models: detection (3.0M) + direction classifier (1.4M) + recognition (5.0M) = 9.4M.Ultra lightweight PP-OCRv2 series models: detection (3.1M) + direction classifier (1.4M) + recognition 8.5M) = 13.0M.PP-OCR series of high-quality pre-trained models, comparable to commercial effects.Find tutorial here；release multi language recognition models, support more than 80 languages recognition especically, the performance of English recognition model is Optimized.
( arxiv paper)Ģ021.8.3 released PaddleOCR v2.2, add a new structured documents analysis toolkit, i.e., PP-Structure, support layout analysis and table recognition (One-key to export chart images to Excel files).Ģ021.4.8 release end-to-end text recognition algorithm PGNet which is published in AAAI 2021. The F-score of PP-OCRv2 is 7% higher than that of PP-OCR mobile. The inference speed of PP-OCRv2 is 220% higher than that of PP-OCR server in CPU device. PaddleOCR R&D team would like to share the key points of PP-OCRv2, at 20:15 pm on September 8th, Live Address.Ģ021.9.7 release PaddleOCR v2.3, PP-OCRv2 is proposed. PaddleOCR aims to create multilingual, awesome, leading, and practical OCR tools that help users train better models and apply them into practice.
0 kommentar(er)
