# PP-OCRv2: Bag of Tricks for Ultra Lightweight OCR System

arXiv: https://arxiv.org/abs/2109.03144 · HTML: https://mlsanity.com/paper/2109.03144/

## Key facts

- Cited by 39 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 22 references that are in the map.
- First posted on arXiv 7 Sep 2021.

## Builds on (most cited first, 14 of 14)

- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [Deep Mutual Learning](https://mlsanity.com/paper/1706.00384/) (2017)
- [Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation](https://mlsanity.com/paper/2012.07177/) (2020)
- [An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition](https://mlsanity.com/paper/1507.05717/) (2015)
- [A Comprehensive Overhaul of Feature Distillation](https://mlsanity.com/paper/1904.01866/) (2019)
- [ICDAR2019 Competition on Scanned Receipt OCR and Information Extraction](https://mlsanity.com/paper/2103.10213/) (2021)
- [Real-time Scene Text Detection with Differentiable Binarization](https://mlsanity.com/paper/1911.08947/) (2019)
- [PP-OCR: A Practical Ultra Lightweight OCR System](https://mlsanity.com/paper/2009.09941/) (2020)
- [ICDAR2019 Robust Reading Challenge on Multi-lingual Scene Text Detection and Recognition -- RRC-MLT-2019](https://mlsanity.com/paper/1907.00945/) (2019)
- [MixConv: Mixed Depthwise Convolutional Kernels](https://mlsanity.com/paper/1907.09595/) (2019)
- [ICDAR2017 Competition on Reading Chinese Text in the Wild (RCTW-17)](https://mlsanity.com/paper/1708.09585/) (2017)
- [Chinese Street View Text: Large-scale Chinese Text Reading with Partially Supervised Learning](https://mlsanity.com/paper/1909.07808/) (2019)

## Built on by (most cited first, 30 of 39)

- [General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model](https://mlsanity.com/paper/2409.01704/) (2024)
- [PaddleOCR 3.0 Technical Report](https://mlsanity.com/paper/2507.05595/) (2025)
- [GlyphControl: Glyph Conditional Control for Visual Text Generation](https://mlsanity.com/paper/2305.18259/) (2023)
- [PP-OCRv3: More Attempts for the Improvement of Ultra Lightweight OCR System](https://mlsanity.com/paper/2206.03001/) (2022)
- [Focus Anywhere for Fine-grained Multi-page Document Understanding](https://mlsanity.com/paper/2405.14295/) (2024)
- [MoAI: Mixture of All Intelligence for Large Language and Vision Models](https://mlsanity.com/paper/2403.07508/) (2024)
- [GlyphDraw: Seamlessly Rendering Text with Intricate Spatial Structures in Text-to-Image Generation](https://mlsanity.com/paper/2303.17870/) (2023)
- [Ocean-OCR: Towards General OCR Application via a Vision-Language Model](https://mlsanity.com/paper/2501.15558/) (2025)
- [Implicit Concept Removal of Diffusion Models](https://mlsanity.com/paper/2310.05873/) (2023)
- [PP-StructureV2: A Stronger Document Analysis System](https://mlsanity.com/paper/2210.05391/) (2022)
- [DocParser: End-to-end OCR-free Information Extraction from Visually Rich Documents](https://mlsanity.com/paper/2304.12484/) (2023)
- [Logics-Parsing Technical Report](https://mlsanity.com/paper/2509.19760/) (2025)
- [From Training-Free to Adaptive: Empirical Insights into MLLMs' Understanding of Detection Information](https://mlsanity.com/paper/2401.17981/) (2024)
- [PP-OCRv6: From 1.5M to 34.5M Parameters, Surpassing Billion-Scale VLMs on OCR Tasks](https://mlsanity.com/paper/2606.13108/) (2026)
- [Rethinking Visual Prompting for Multimodal Large Language Models with External Knowledge](https://mlsanity.com/paper/2407.04681/) (2024)
- [R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts](https://mlsanity.com/paper/2502.20395/) (2025)
- [Computational Optics for Mobile Terminals in Mass Production](https://mlsanity.com/paper/2305.05886/) (2023)
- [What Large Language Models Bring to Text-rich VQA?](https://mlsanity.com/paper/2311.07306/) (2023)
- [VLsI: Verbalized Layers-to-Interactions from Large to Small Vision Language Models](https://mlsanity.com/paper/2412.01822/) (2024)
- [DianJin-OCR-R1: Enhancing OCR Capabilities via a Reasoning-and-Tool Interleaved Vision-Language Model](https://mlsanity.com/paper/2508.13238/) (2025)
- [Bridging the Gap Between End-to-End and Two-Step Text Spotting](https://mlsanity.com/paper/2404.04624/) (2024)
- [Self-distillation Regularized Connectionist Temporal Classification Loss for Text Recognition: A Simple Yet Effective Approach](https://mlsanity.com/paper/2308.08806/) (2023)
- [Hallucination in Multimodal Foundation Models: A Survey on Causes, Corrections, and Evaluations](https://mlsanity.com/paper/2410.15359/) (2024)
- [PatentLMM: Large Multimodal Model for Generating Descriptions for Patent Figures](https://mlsanity.com/paper/2501.15074/) (2025)
- [Table Structure Recognition with Conditional Attention](https://mlsanity.com/paper/2203.03819/) (2022)
- [Masking Teacher and Reinforcing Student for Distilling Vision-Language Models](https://mlsanity.com/paper/2512.22238/) (2025)
- [TC-OCR: TableCraft OCR for Efficient Detection & Recognition of Table Structure & Content](https://mlsanity.com/paper/2404.10305/) (2024)
- [UI-UG: A Unified MLLM for UI Understanding and Generation](https://mlsanity.com/paper/2509.24361/) (2025)
- [Selective Vision is the Challenge for Visual Reasoning: A Benchmark for Visual Argument Understanding](https://mlsanity.com/paper/2406.18925/) (2024)
- [Anonymization of Documents for Law Enforcement with Machine Learning](https://mlsanity.com/paper/2501.07334/) (2025)

## Abstract (arXiv)

Optical Character Recognition (OCR) systems have been widely used in various of application scenarios. Designing an OCR system is still a challenging task. In previous work, we proposed a practical ultra lightweight OCR system (PP-OCR) to balance the accuracy against the efficiency. In order to improve the accuracy of PP-OCR and keep high efficiency, in this paper, we propose a more robust OCR system, i.e. PP-OCRv2. We introduce bag of tricks to train a better text detector and a better text recognizer, which include Collaborative Mutual Learning (CML), CopyPaste, Lightweight CPUNetwork (LCNet), Unified-Deep Mutual Learning (U-DML) and Enhanced CTCLoss. Experiments on real data show that the precision of PP-OCRv2 is 7% higher than PP-OCR under the same inference cost. It is also comparable to the server models of the PP-OCR which uses ResNet series as backbones. All of the above mentioned models are open-sourced and the code is available in the GitHub repository PaddleOCR which is powered by PaddlePaddle.