README_en.md

🌟 Features | >🌐 Online Experience｜🚀 Quick Start | 📖 Documentation | 🔥Capabilities | 📋 Models

🇨🇳 Simplified Chinese | 🇬🇧 English

🔍 Introduction

PaddleX 3.0 is a low-code development tool for AI models built on the PaddlePaddle framework. It integrates numerous ready-to-use pre-trained models, enabling full-process development from model training to inference, supporting a variety of mainstream hardware both domestic and international, and aiding AI developers in industrial practice.

🌟 Why PaddleX ?

🎨 Rich Models One-click Call: Integrate over 200 PaddlePaddle models covering multiple key areas such as OCR, object detection, and time series forecasting into 33 pipelines. Experience the model effects quickly through easy Python API calls. Also supports 39 modules for easy model combination use by developers.

🚀 High Efficiency and Low barrier of entry: Achieve model full-process development based on graphical interfaces and unified commands, creating 8 featured model pipelines that combine large and small models, semi-supervised learning of large models, and multi-model fusion, greatly reducing the cost of iterating models.

🌐 Flexible Deployment in Various Scenarios: Support various deployment methods such as high-performance inference, serving, and lite deployment to ensure efficient operation and rapid response of models in different application scenarios.

🔧 Efficient Support for Mainstream Hardware: Support seamless switching of various mainstream hardware such as NVIDIA GPUs, Kunlun XPU, Ascend NPU, and Cambricon MLU to ensure efficient operation.

📣 Recent Updates

🔥🔥 2025.10.16, PaddleX v3.3.0 Released

• Added support for inference and deployment of PaddleOCR-VL and PP-OCRv5 multilingual models.

🔥🔥 2025.8.20, PaddleX v3.2.0 Released

• Deployment Capability Upgrades:

  • Fully supports PaddlePaddle framework versions 3.1.0 and 3.1.1.
  • High-performance inference supports CUDA 12, with backend options including Paddle Inference and ONNX Runtime.
  • High-stability serving solution is fully open-sourced, enabling users to customize Docker images and SDKs as needed.
  • High-stability serving solution supports invocation via manually constructed HTTP requests, allowing client applications to be developed in any programming language.

• Key Model Additions:

  • Added training, inference, and deployment support for PP-OCRv5 English, Thai, and Greek recognition models. The PP-OCRv5 English model delivers an 11% improvement over the main PP-OCRv5 model in English scenarios, with the Thai model achieving an accuracy of 82.68% and the Greek model 89.28%.

• Benchmark Enhancements:

  • All pipelines support fine-grained benchmarking, enabling the measurement of end-to-end inference time as well as per-layer and per-module latency data to assist with performance analysis.
  • Added key metrics such as inference latency and memory usage for commonly used configurations on mainstream hardware to the documentation, providing deployment reference for users.

• Bug Fixes:

  • Fixed an issue where invalid input image file formats could cause recursive calls.
  • Resolved ineffective parameter settings for chart recognition, seal recognition, and document pre-processing in the configuration files for the PP-DocTranslation and PP-StructureV3 pipelines.
  • Fixed an issue where PDF files were not properly closed after inference.

• Other Updates:

  • Added support for Windows users with NVIDIA 50-series graphics cards; users can install the corresponding PaddlePaddle framework version as per the installation guide.
  • The PP-OCR model series now supports returning coordinates for individual characters.
  • The model_name parameter in PaddlePredictorOption has been moved to PaddleInfer, improving usability.
  • Refactored the official model download logic, with new support for multiple model hosting platforms such as AIStudio and ModelScope.

🔥🔥 2025.6.28, PaddleX v3.1.0 Released

• Key Models:
  • Added PP-OCRv5 Multilingual Text Recognition Model, supporting training and inference workflows for text recognition in 37 languages, including French, Spanish, Portuguese, Russian, Korean, and more. Average precision increased by over 30%.
  • Upgraded PP-Chart2Table model in PP-StructureV3. Chart-to-table conversion capability further improved, with RMS-F1 on our internal evaluation set increased by 9.36 percentage points (71.24% -> 80.60%).
• Key Pipelines:
  • Added document translation pipeline PP-DocTranslation based on PP-StructureV3 and ERNIE 4.5 Turbo. Supports translation of Markdown documents, various complex-layout PDF documents, and document images, with results saved as Markdown format documents.

🔥🔥 2025.5.20: PaddleX v3.0.0 Released

Core upgrades are as follows:

• Rich Model Library:

  • Extensive Model Coverage: PaddleX 3.0 includes 270+ models, covering diverse scenarios such as image/video classification/detection/segmentation, OCR, speech recognition, time series analysis, and more.
  • Mature Solutions: Built on this robust model library, PaddleX 3.0 offers critical and production-ready AI solutions, including general document parsing, key information extraction, document understanding, table recognition, and general image recognition.

• Unified Inference API & Enhanced Deployment Capabilities:

  • Standardized Inference Interface: Reduces API fragmentation across model types, lowering the learning curve for users and accelerating enterprise adoption.
  • Multi-Model Composition: Complex tasks can be efficiently tackled by combining different models, achieving synergistic performance (1+1>2).
  • Upgraded Deployment: Unified commands now manage deployments for diverse models, supporting multi-GPU inference and multi-instance serving deployments.

• Full Compatibility with PaddlePaddle Framework 3.0:

  • Leveraging New Paddle 3.0 Features:
    • Compiler-accelerated training: Enable by appending -o Global.dy2st=True to training commands. Most GPU-based models see >10% speed gains, with some exceeding 30%.
    • Inference upgrades: Full adaptation to Paddle 3.0’s Program Intermediate Representation (PIR) enhances flexibility and compatibility. Static graph models now use xxx.json instead of xxx.pdmodel.
  • ONNX Model Support: Seamless format conversion via the Paddle2ONNX plugin.

• Flagship Capabilities:

  • PP-OCRv5: Powers multi-hardware inference, multi-backend support, and serving deployments for this industry-leading OCR system.
  • PP-StructureV3: Orchestrates 15+ models in hybrid (serial/parallel) pipelines, achieving SOTA accuracy on OmniDocBench.
  • PP-ChatOCRv4: Integrates with PP-DocBee2 and ERNIE 4.5Turbo, boosting key information extraction accuracy by 15.7 percentage points over the previous generation.

• Multi-Hardware Support:

  • Broad Compatibility: Training and inference supported on NVIDIA, Intel, Apple M-series, Kunlunxin, Ascend, Cambricon, Hygon, Enflame, and more.
  • Ascend-Optimized: 200+ fully adapted models, including 21 OM-accelerated inference models, plus key solutions like PP-OCRv5 and PP-StructureV3.
  • Kunlunxin-Optimized: Critical classification, detection, and OCR models (including PP-OCRv5) are fully supported.

🔠 Explanation of Pipeline

PaddleX is dedicated to achieving pipeline-level model training, inference, and deployment. A pipeline refers to a series of predefined development processes for specific AI tasks, which includes a combination of single models (single-function modules) capable of independently completing a certain type of task.

📊 What can PaddleX do？

All pipelines of PaddleX support online experience on AI Studio and local fast inference. You can quickly experience the effects of each pre-trained pipeline. If you are satisfied with the effects of the pre-trained pipeline, you can directly perform high-performance inference / serving / edge deployment on the pipeline. If not satisfied, you can also Custom Development to improve the pipeline effect. For the complete pipeline development process, please refer to the PaddleX pipeline Development Tool Local Use Tutorial.

In addition, PaddleX provides developers with a full-process efficient model training and deployment tool based on a cloud-based GUI. Developers do not need code development, just need to prepare a dataset that meets the pipeline requirements to quickly start model training. For details, please refer to the tutorial "Developing Industrial-level AI Models with Zero Barrier".

Pipeline	Online Experience	Local Inference	High-Performance Inference	Serving	On-Device Deployment	Custom Development	Zero-Code Development On AI Studio
OCR	Link	✅	✅	✅	✅	✅	✅
PP-ChatOCRv3	Link	✅	✅	✅	🚧	✅	✅
PP-ChatOCRv4	Link	✅	✅	✅	🚧	✅	✅
Table Recognition	Link	✅	✅	✅	🚧	✅	✅
Object Detection	Link	✅	✅	✅	✅	✅	✅
Instance Segmentation	Link	✅	✅	✅	🚧	✅	✅
Image Classification	Link	✅	✅	✅	✅	✅	✅
Semantic Segmentation	Link	✅	✅	✅	✅	✅	✅
Time Series Forecasting	Link	✅	✅	✅	🚧	✅	✅
Time Series Anomaly Detection	Link	✅	✅	✅	🚧	✅	✅
Time Series Classification	Link	✅	✅	✅	🚧	✅	✅
Small Object Detection	Link	✅	✅	✅	🚧	✅	✅
Multi-label Image Classification	Link	✅	✅	✅	🚧	✅	✅
Pedestrian Attribute Recognition	Link	✅	✅	✅	🚧	✅	✅
Vehicle Attribute Recognition	Link	✅	✅	✅	🚧	✅	✅
Formula Recognition	Link	✅	✅	✅	🚧	✅	✅
Seal Recognition	Link	✅	✅	✅	🚧	✅	✅
Image Anomaly Detection	🚧	✅	✅	✅	🚧	✅	🚧
Human Keypoint Detection	🚧	✅	✅	✅	🚧	✅	🚧
Open Vocabulary Detection	🚧	✅	✅	✅	🚧	🚧	🚧
Open Vocabulary Segmentation	🚧	✅	✅	✅	🚧	🚧	🚧
Rotated Object Detection	🚧	✅	✅	✅	🚧	✅	🚧
3D Bev Detection	🚧	✅	✅	✅	🚧	✅	🚧
Table Recognition v2	Link	✅	✅	✅	🚧	✅	✅
Layout Parsing	🚧	✅	✅	✅	🚧	✅	🚧
PP-StructureV3	Link	✅	✅	✅	🚧	🚧	✅
Document Image Preprocessing	🚧	✅	✅	✅	🚧	✅	🚧
Image Recognition	🚧	✅	✅	✅	🚧	✅	🚧
Face Recognition	🚧	✅	✅	✅	🚧	✅	🚧
Multilingual Speech Recognition	🚧	✅	🚧	✅	🚧	🚧	🚧
Video Classification	🚧	✅	✅	✅	🚧	✅	🚧
Video Detection	🚧	✅	✅	✅	🚧	✅	🚧
Document Understanding	🚧	✅	🚧	✅	🚧	🚧	🚧

❗Note: The above capabilities are implemented based on GPU/CPU. PaddleX can also perform local inference and custom development on mainstream hardware such as Kunlunxin, Ascend, Cambricon, and HYGON. The table below details the support status of the pipelines. For specific supported model lists, please refer to the Model List (Kunlunxin XPU)/Model List (Ascend NPU)/Model List (Cambricon MLU)/Model List (HYGON DCU). We are continuously adapting more models and promoting the implementation of high-performance and serving on mainstream hardware.

🔥🔥 Support for Domestic Hardware Capabilities

Pipeline	Ascend 910B	Kunlunxin R200/R300	Cambricon MLU370X8	HYGON Z100
OCR	✅	✅	✅	🚧
Table Recognition	✅	🚧	🚧	🚧
Object Detection	✅	✅	✅	🚧
Instance Segmentation	✅	🚧	✅	🚧
Image Classification	✅	✅	✅	✅
Semantic Segmentation	✅	✅	✅	✅
Time Series Forecasting	✅	✅	✅	🚧
Time Series Anomaly Detection	✅	🚧	🚧	🚧
Time Series Classification	✅	🚧	🚧	🚧
Multi-label Image Classification	✅	🚧	🚧	✅
Pedestrian Attribute Recognition	✅	🚧	🚧	🚧
Vehicle Attribute Recognition	✅	🚧	🚧	🚧
Image Recognition	✅	🚧	✅	✅
Seal Recognition	✅	🚧	🚧	🚧
Image Anomaly Detection	✅	✅	✅	✅
Face Recognition	✅	✅	✅	✅

⏭️ Quick Start

🛠️ Installation

❗Before installing PaddleX, please ensure that you have a basic Python runtime environment (Note: Currently supports Python 3.8 to Python 3.13). The PaddleX 3.0.x version depends on PaddlePaddle version 3.0.0 and above. Please make sure the version compatibility is maintained before use.

• Installing PaddlePaddle

# CPU
python -m pip install paddlepaddle==3.3.0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/

# gpu，requires GPU driver version ≥450.80.02 (Linux) or ≥452.39 (Windows)
 python -m pip install paddlepaddle-gpu==3.3.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/

# gpu，requires GPU driver version ≥550.54.14 (Linux) or ≥550.54.14 (Windows)
 python -m pip install paddlepaddle-gpu==3.3.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu126/

❗No need to focus on the CUDA version on the physical machine, only the GPU driver version needs attention. For more information on PaddlePaddle Wheel versions, please refer to the PaddlePaddle Official Website.

• Installing PaddleX

pip install "paddlex[base]"

❗For more installation methods, refer to the PaddleX Installation Guide.

💻 CLI Usage

One command can quickly experience the pipeline effect, the unified CLI format is:

paddlex --pipeline [Pipeline Name] --input [Input Image] --device [Running Device]

Each Pipeline in PaddleX corresponds to specific parameters, which you can view in the respective Pipeline documentation for detailed explanations. Each Pipeline requires specifying three necessary parameters:

• pipeline: The name of the Pipeline or the configuration file of the Pipeline
• input: The local path, directory, or URL of the input file (e.g., an image) to be processed
• device: The hardware device and its index to use (e.g., gpu:0 indicates using the 0th GPU), or you can choose to use NPU (npu:0), XPU (xpu:0), CPU (cpu), etc.

For example, using the OCR pipeline:

paddlex --pipeline OCR \
        --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_ocr_002.png \
        --use_doc_orientation_classify False \
        --use_doc_unwarping False \
        --use_textline_orientation False \
        --save_path ./output \
        --device gpu:0

👉 Click to view the running result

{'res': {'input_path': 'general_ocr_002.png', 'page_index': None, 'model_settings': {'use_doc_preprocessor': False, 'use_textline_orientation': False}, 'doc_preprocessor_res': {'input_path': None, 'model_settings': {'use_doc_orientation_classify': True, 'use_doc_unwarping': False}, 'angle': 0},'dt_polys': [array([[ 3, 10],
       [82, 10],
       [82, 33],
       [ 3, 33]], dtype=int16), ...], 'text_det_params': {'limit_side_len': 960, 'limit_type': 'max', 'thresh': 0.3, 'box_thresh': 0.6, 'unclip_ratio': 2.0}, 'text_type': 'general', 'textline_orientation_angles': [-1, ...], 'text_rec_score_thresh': 0.0, 'rec_texts': ['www.99*', ...], 'rec_scores': [0.8980069160461426,  ...], 'rec_polys': [array([[ 3, 10],
       [82, 10],
       [82, 33],
       [ 3, 33]], dtype=int16), ...], 'rec_boxes': array([[  3,  10,  82,  33], ...], dtype=int16)}}

The visualization result is as follows:

To use the command line for other pipelines, simply adjust the pipeline parameter to the name of the corresponding pipeline and modify the parameters accordingly. Below are the commands for each pipeline:

👉 More CLI usage for pipelines

Pipeline Name	Command
General Image Classification	paddlex --pipeline image_classification --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_image_classification_001.jpg --device gpu:0 --save_path ./output/
General Object Detection	paddlex --pipeline object_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_object_detection_002.png --threshold 0.5 --save_path ./output/ --device gpu:0
General Instance Segmentation	paddlex --pipeline instance_segmentation --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_instance_segmentation_004.png --threshold 0.5 --save_path ./output --device gpu:0
General Semantic Segmentation	paddlex --pipeline semantic_segmentation --input https://paddle-model-ecology.bj.bcebos.com/paddlex/PaddleX3.0/application/semantic_segmentation/makassaridn-road_demo.png --target_size -1 --save_path ./output --device gpu:0
Image Multi-label Classification	paddlex --pipeline image_multilabel_classification --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_image_classification_001.jpg --save_path ./output --device gpu:0
Small Object Detection	paddlex --pipeline small_object_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/small_object_detection.jpg --threshold 0.5 --save_path ./output --device gpu:0
Image Anomaly Detection	paddlex --pipeline anomaly_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/uad_grid.png --save_path ./output --device gpu:0
Pedestrian Attribute Recognition	paddlex --pipeline pedestrian_attribute_recognition --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/pedestrian_attribute_002.jpg --save_path ./output/ --device gpu:0
Vehicle Attribute Recognition	paddlex --pipeline vehicle_attribute_recognition --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/vehicle_attribute_002.jpg --save_path ./output/ --device gpu:0
3D Multi-modal Fusion Detection	paddlex --pipeline 3d_bev_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/det_3d/demo_det_3d/nuscenes_demo_infer.tar --device gpu:0 --save_path ./output/
Human Keypoint Detection	paddlex --pipeline human_keypoint_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/keypoint_detection_001.jpg --det_threshold 0.5 --save_path ./output/ --device gpu:0
Open Vocabulary Detection	paddlex --pipeline open_vocabulary_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/open_vocabulary_detection.jpg --prompt "bus . walking man . rearview mirror ." --thresholds "{'text_threshold': 0.25, 'box_threshold': 0.3}" --save_path ./output --device gpu:0
Open Vocabulary Segmentation	paddlex --pipeline open_vocabulary_segmentation --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/open_vocabulary_segmentation.jpg --prompt_type box --prompt "[[112.9,118.4,513.8,382.1],[4.6,263.6,92.2,336.6],[592.4,260.9,607.2,294.2]]" --save_path ./output --device gpu:0
Rotated Object Detection	paddlex --pipeline rotated_object_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/rotated_object_detection_001.png --threshold 0.5 --save_path ./output --device gpu:0
General OCR	paddlex --pipeline OCR --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_ocr_002.png --use_doc_orientation_classify False --use_doc_unwarping False --use_textline_orientation False --save_path ./output --device gpu:0
Document Image Preprocessor	paddlex --pipeline doc_preprocessor --input https://paddle-model-ecology.bj.bcebos.com/paddlex/demo_image/doc_test_rotated.jpg --use_doc_orientation_classify True --use_doc_unwarping True --save_path ./output --device gpu:0
General Table Recognition	paddlex --pipeline table_recognition --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/table_recognition.jpg --save_path ./output --device gpu:0
General Table Recognition v2	paddlex --pipeline table_recognition_v2 --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/table_recognition.jpg --save_path ./output --device gpu:0
General Layout Parsing	paddlex --pipeline layout_parsing --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/demo_paper.png --use_doc_orientation_classify False --use_doc_unwarping False --use_textline_orientation False --save_path ./output --device gpu:0
General Layout Parsing v2	paddlex --pipeline PP-StrucutrV3 --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/pp_structure_v3_demo.png --use_doc_orientation_classify False --use_doc_unwarping False --use_textline_orientation False --save_path ./output --device gpu:0
Formula Recognition	paddlex --pipeline formula_recognition --input https://paddle-model-ecology.bj.bcebos.com/paddlex/demo_image/general_formula_recognition.png --use_layout_detection True --use_doc_orientation_classify False --use_doc_unwarping False --layout_threshold 0.5 --layout_nms True --layout_unclip_ratio 1.0 --layout_merge_bboxes_mode large --save_path ./output --device gpu:0
Seal Text Recognition	paddlex --pipeline seal_recognition --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/seal_text_det.png --use_doc_orientation_classify False --use_doc_unwarping False --device gpu:0 --save_path ./output
Time Series Forecasting	paddlex --pipeline ts_forecast --input https://paddle-model-ecology.bj.bcebos.com/paddlex/ts/demo_ts/ts_fc.csv --device gpu:0 --save_path ./output
Time Series Anomaly Detection	paddlex --pipeline ts_anomaly_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/ts/demo_ts/ts_ad.csv --device gpu:0 --save_path ./output
Time Series Classification	paddlex --pipeline ts_classification --input https://paddle-model-ecology.bj.bcebos.com/paddlex/ts/demo_ts/ts_cls.csv --device gpu:0 --save_path ./output
Multilingual Speech Recognition	paddlex --pipeline multilingual_speech_recognition --input https://paddlespeech.bj.bcebos.com/PaddleAudio/zh.wav --save_path ./output --device gpu:0
General Video Classification	paddlex --pipeline video_classification --input https://paddle-model-ecology.bj.bcebos.com/paddlex/videos/demo_video/general_video_classification_001.mp4 --topk 5 --save_path ./output --device gpu:0
General Video Detection	paddlex --pipeline video_detection --input https://paddle-model-ecology.bj.bcebos.com/paddlex/videos/demo_video/HorseRiding.avi --device gpu:0 --save_path ./output

📝 Python Script Usage

A few lines of code can complete the quick inference of the pipeline, the unified Python script format is as follows:

from paddlex import create_pipeline

pipeline = create_pipeline(pipeline=[Pipeline Name])
output = pipeline.predict([Input Image Name])
for res in output:
    res.print()
    res.save_to_img("./output/")
    res.save_to_json("./output/")

The following steps are executed:

• create_pipeline() instantiates the pipeline object
• Passes the image and calls the predict() method of the pipeline object for inference prediction
• Processes the prediction results

To use the Python script for other pipelines, simply adjust the pipeline parameter in the create_pipeline() method to the name of the corresponding pipeline and modify the parameters accordingly. Below are the parameter names and detailed usage explanations for each pipeline:

👉 More Python script usage for pipelines

pipeline Name	Corresponding Parameter	Detailed Explanation
PP-ChatOCRv4-doc	PP-ChatOCRv4-doc	PP-ChatOCRv4-doc Pipeline Python Script Usage Instructions
PP-ChatOCRv3-doc	PP-ChatOCRv3-doc	PP-ChatOCRv3-doc Pipeline Python Script Usage Instructions
Image Classification	image_classification	Image Classification Pipeline Python Script Usage Instructions
Object Detection	object_detection	Object Detection Pipeline Python Script Usage Instructions
Instance Segmentation	instance_segmentation	Instance Segmentation Pipeline Python Script Usage Instructions
Semantic Segmentation	semantic_segmentation	Semantic Segmentation Pipeline Python Script Usage Instructions
Image Multi-Label Classification	multilabel_classification	Image Multi-Label Classification Pipeline Python Script Usage Instructions
Small Object Detection	small_object_detection	Small Object Detection Pipeline Python Script Usage Instructions
Image Anomaly Detection	image_classification	Image Anomaly Detection Pipeline Python Script Usage Instructions
Image Recognition	PP-ShiTuV2	Image Recognition Pipeline Python Script Usage Instructions
Face Recognition	face_recognition	Face Recognition Pipeline Python Script Usage Instructions
Pedestrian Attribute Recognition	pedestrian_attribute	Pedestrian Attribute Recognition Pipeline Python Script Usage Instructions
Vehicle Attribute Recognition	vehicle_attribute	Vehicle Attribute Recognition Pipeline Python Script Usage Instructions
3D Multi-modal Fusion Detection	3d_bev_detection	Instructions for Using the 3D Multi-modal Fusion Detection Pipeline Python Script
Human Keypoint Detection	human_keypoint_detection	Instructions for Using the Human Keypoint Detection Pipeline Python Script
Open Vocabulary Detection	open_vocabulary_detection	Instructions for Using the Open Vocabulary Detection Pipeline Python Script
Open Vocabulary Segmentation	open_vocabulary_segmentation	Instructions for Using the Open Vocabulary Segmentation Pipeline Python Script
Rotated Object Detection	rotated_object_detection	Instructions for Using the Rotated Object Detection Pipeline Python Script
OCR	OCR	Instructions for Using the General OCR Pipeline Python Script
Document Image Preprocessing	doc_preprocessor	Instructions for Using the Document Image Preprocessing Pipeline Python Script
General Table Recognition	table_recognition	Instructions for Using the General Table Recognition Pipeline Python Script
General Table Recognition v2	table_recognition_v2	Instructions for Using the General Table Recognition v2 Pipeline Python Script
General Layout Parsing	layout_parsing	Instructions for Using the General Layout Parsing Pipeline Python Script
PP-StructureV3	PP-StructureV3	Instructions for Using the General Layout Parsing v2 Pipeline Python Script
Formula Recognition	formula_recognition	Instructions for Using the Formula Recognition Pipeline Python Script
Seal Text Recognition	seal_recognition	Instructions for Using the Seal Text Recognition Pipeline Python Script
Time Series Forecasting	ts_forecast	Instructions for Using the Time Series Forecasting Pipeline Python Script
Time Series Anomaly Detection	ts_anomaly_detection	Instructions for Using the Time Series Anomaly Detection Pipeline Python Script
Time Series Classification	ts_classification	Instructions for Using the Time Series Classification Pipeline Python Script
Multilingual Speech Recognition	multilingual_speech_recognition	Instructions for Using the Multilingual Speech Recognition Pipeline Python Script
General Video Classification	video_classification	Instructions for Using the General Video Classification Pipeline Python Script
General Video Detection	video_detection	Instructions for Using the General Video Detection Pipeline Python Script
Document Understanding	doc_understanding	Instructions for Using the Document Understanding Pipeline Python Script

📖 Documentation

⬇️ Installation

• 📦 PaddlePaddle Installation
• 📦 PaddleX Installation

🔥 Pipeline Usage

• 📑 PaddleX Pipeline Usage Overview

• 📝 Information Extraction

  • 📄 PP-ChatOCRv3 Pipeline Tutorial
  • 📄 PP-ChatOCRv4 Pipeline Tutorial

• 🔍 OCR

  • 📜 OCR Pipeline Tutorial
  • 📊 Table Recognition Pipeline Tutorial
  • 🗂️ Table Recognition v2 Pipeline Tutorial
  • 📄 Layout Parsing Pipeline Tutorial
  • 🗞️ PP-StructureV3 Pipeline Tutorial
  • 📐 Formula Recognition Pipeline Tutorial
  • 📝 Seal Recognition Pipeline Tutorial
  • 🖌️ Document Image Preprocessing

• 🎥 Computer Vision

  • 🖼️ Image Classification Pipeline Tutorial
  • 🎯 Object Detection Pipeline Tutorial
  • 📋 Instance Segmentation Pipeline Tutorial
  • 🗣️ Semantic Segmentation Pipeline Tutorial
  • 🏷️ Multi-label Image Classification Pipeline Tutorial
  • 🔍 Small Object Detection Pipeline Tutorial
  • 🖼️ Image Anomaly Detection Pipeline Tutorial
  • 🌐 3D Bev Detection Pipeline Tutorial
  • 🔍 Human Keypoint Detection Pipeline Tutorial
  • 📚 Open Vocabulary Detection Pipeline Tutorial
  • 🎨 Open Vocabulary Segmentation Pipeline Tutorial
  • 🔄 Rotated Object Detection Pipeline Tutorial
  • 🖼️ Image Recognition Pipeline Tutorial
  • 🚶‍♀️ Pedestrian Attribute Recognition Pipeline Tutorial
  • 🚗 Vehicle Attribute Recognition Pipeline Tutorial
  • 🆔 Face Recognition Pipeline Tutorial

• ⏱️ Time Series Analysis

  • 📈 Time Series Forecasting Pipeline Tutorial
  • 📉 Time Series Anomaly Detection Pipeline Tutorial
  • 🕒 Time Series Classification Pipeline Tutorial

• 🎤 Speech Recognition

  • 🌐 Multilingual Speech Recognition Pipeline Usage Tutorial
• 🎥 Video Recognition

  • 📈 General Video Classification Pipeline Usage Tutorial
  • 🔍 General Video Detection Pipeline Usage Tutorial
• 🌐 Multimodal Vision-Language Model

  • 📝 Doc Understanding Pipeline Usage Tutorial
• 🔧 Related Instructions

  • 🖥️ PaddleX pipeline Command Line Instruction
  • 📝 PaddleX pipeline Python Script Instruction
  • 🔎 Line-Parallel Inference

⚙️ Module Usage

• 🔍 OCR

  • 📝 Text Detection Module Tutorial
  • 🔖 Seal Text Detection Module Tutorial
  • 🔠 Text Recognition Module Tutorial
  • 🗺️ Layout Parsing Module Tutorial
  • 📊 Table Structure Recognition Module Tutorial
  • 📊 Table Cell Detection Module Tutorial
  • 📈 Table Classification Module Tutorial
  • 📄 Document Image Orientation Classification Tutorial
  • 🔧 Document Image Unwarp Module Tutorial
  • 📝 Text Line Orientation Classification Module Tutorial
  • 📐 Formula Recognition Module Tutorial

• 🖼️ Image Classification

  • 📂 Image Classification Module Tutorial
  • 🏷️ Multi-label Image Classification Module Tutorial
  • 👤 Pedestrian Attribute Recognition Module Tutorial
  • 🚗 Vehicle Attribute Recognition Module Tutorial

• 🏞️ Image Features

  • 🔗 Image Feature Module Tutorial
  • 😁 Face_Feature Module Tutorial

• 🎯 Object Detection

  • 🎯 Object Detection Module Tutorial
  • 📏 Small Object Detection Module Tutorial
  • 🧑‍🤝‍🧑 Face Detection Module Tutorial
  • 🔍 Mainbody Detection Module Tutorial
  • 🚶 Pedestrian Detection Module Tutorial
  • 🚶‍♂️ Human Keypoint Detection Module Usage Tutorial
  • 🌐 Open-Vocabulary Object Detection Module Usage Tutorial

• 🖼️ Image Segmentation

  • 🗺️ Semantic Segmentation Module Tutorial
  • 🔍 Instance Segmentation Module Tutorial
  • 🚨 Image Anomaly Detection Module Tutorial
  • 🌐 Open-Vocabulary Segmentation Module Tutorial

• ⏱️ Time Series Analysis

  • 📈 Time Series Forecasting Module Tutorial
  • 🚨 Time Series Anomaly Detection Module Tutorial
  • 🕒 Time Series Classification Module Tutorial

• 🎤 Speech Recognition

  • 🌐 Multilingual Speech Recognition Module Usage Tutorial

• 📦 3D

  • 📦 3D Multimodal Fusion Detection Module Usage Tutorial

• 🌐 Multimodal Vision-Language Model

  • 📝 Document Vision-Language Model Usage Tutorial
  • 📈 Chart Parsing Module Usage Tutorial

• 📄 Related Instructions

  • 📝 PaddleX Single Model Python Script Instruction
  • 📝 PaddleX General Model Configuration File Parameter Instruction
  • 📝 PaddleX Time Series Task Model Configuration File Parameter Instruction
  • 📝 PaddleX 3D Task Model Configuration File Parameter Instruction
  • 📝 Model Inference Benchmark

🏗️ Pipeline Deployment

• 🚀 PaddleX High-Performance Inference Guide
• 🖥️ PaddleX Serving Guide
• 📱 PaddleX On-Device Deployment Guide
• 🌐 Installation and Usage of the Paddle2ONNX Plugin

🖥️ Multi-Hardware Usage

• 🔧 Multi-Hardware Usage Guide
• 🖲️ DCU Paddle Installation
• 🔲 MLU Paddle Installation
• 💻 NPU Paddle Installation
• 🔌 XPU Paddle Installation
• 📱 GCU Paddle Installation

📊 Data Annotation Tutorials

• 💻 计算机视觉

  • 📂 Image Classification Task Module
  • 📂 Image Feature Task Module
  • 📂 Instance Segmentation Task Module
  • 📂 Multi-Label Classification Task Module
  • 📂 Object Detection Task Module
  • 📂 Semantic Segmentation Task Module

• 🔍 OCR

  • 📊 Table Structure Recognition Task Module
  • 📰 Text Detection/Text Recognition Task Module

• 📉 时序分析

  • 📈 Time Series Anomaly Detection Task Module
  • 📉Time Series Classification Task Module
  • 🕜 Time Series Forecasting Task Module

📑 Pipeline List

• 🖲️ PaddleX Pipelines (CPU/GPU)
• 🔲 PaddleX Pipelines (DCU)
• 💻 PaddleX Pipelines (MLU)
• 🔌 PaddleX Pipelines (NPU)
• 📱 PaddleX Pipelines (XPU)

📄 Model List

• 🖲️ PaddleX Model List (CPU/GPU)
• 🔲 PaddleX Model List (Hygon DCU)
• 💻 PaddleX Model List (Cambricon MLU)
• 🔌 PaddleX Model List (Huawei Ascend NPU)
• 📱 PaddleX Model List (Kunlun XPU)
• 📺 PaddleX Model List (Enflame GCU)

📝 Tutorials & Examples

• 📑 PP-ChatOCRv3 Model Line —— Paper Document Information Extract Tutorial
• 📑 PP-ChatOCRv3 Model Line —— Seal Information Extract Tutorial
• 📑 Document Scene Information Extraction v3 (PP-ChatOCRv3_doc) -- DeepSeek Edition
• 🚗 General OCR Model Line —— License Plate Recognition Tutorial
• ✍️ General OCR Model Line —— Handwritten Chinese Character Recognition Tutorial
• 🔍 Practical Guide to Formula Recognition Model Production Line
• 💻 Layout Detection Model Pipeline Tutorial —— Large Model Training Data Construction Tutorial
• 😊 Face Recognition Pipeline —— Cartoon Face Recognition Tutorial
• 🖼️ General Image Classification Model Line —— Garbage Classification Tutorial
• 🧩 General Instance Segmentation Model Line —— Remote Sensing Image Instance Segmentation Tutorial
• 👥 General Object Detection Model Line —— Pedestrian Fall Detection Tutorial
• 👗 General Object Detection Model Line —— Fashion Element Detection Tutorial
• 🗣️ General Semantic Segmentation Model Line —— Road Line Segmentation Tutorial
• 🛠️ Time Series Anomaly Detection Model Line —— Equipment Anomaly Detection Application Tutorial
• 🎢 Time Series Classification Model Line —— Heartbeat Monitoring Time Series Data Classification Application Tutorial
• 🔋 Time Series Forecasting Model Line —— Long-term Electricity Consumption Forecasting Application Tutorial
• 🔧 Pipeline Deployment Tutorial

🤔 FAQ

For answers to some common questions about our project, please refer to the FAQ. If your question has not been answered, please feel free to raise it in Issues.

💬 Discussion

We warmly welcome and encourage community members to raise questions, share ideas, and feedback in the Discussions section. Whether you want to report a bug, discuss a feature request, seek help, or just want to keep up with the latest project news, this is a great platform.

📄 License

The release of this project is licensed under the Apache 2.0 license.