PP-OCR¶
Class: PPOCRBlockV1
Source: inference.core.workflows.core_steps.models.foundation.pp_ocr.v1.PPOCRBlockV1
Retrieve the characters in an image using PP-OCR (PaddleOCR) Optical Character Recognition (OCR).
This block returns the text within an image.
You may want to use this block in combination with a detections-based block (i.e. ObjectDetectionBlock). An object detection model could isolate specific regions from an image (i.e. a shipping container ID in a logistics use case) for further processing. You can then use a DynamicCropBlock to crop the region of interest before running OCR.
Using a detections model then cropping detections allows you to isolate your analysis on particular regions of an image.
Type identifier¶
Use the following identifier in step "type" field: roboflow_core/pp_ocr@v1to add the block as
as step in your workflow.
Properties¶
| Name | Type | Description | Refs |
|---|---|---|---|
name |
str |
Unique name of step in workflows. | ❌ |
text_detection |
str |
Text detection model size. Set to none to disable detection and run recognition on each input image as a single, pre-cropped text line.. |
❌ |
text_recognition |
str |
Text recognition model size. Set to none to disable recognition and return detected text boxes only, without transcribed text.. |
❌ |
The Refs column marks possibility to parametrise the property with dynamic values available
in workflow runtime. See Bindings for more info.
Available Connections¶
Compatible Blocks
Check what blocks you can connect to PP-OCR in version v1.
- inputs:
Keypoint Visualization,Morphological Transformation,Blur Visualization,Bounding Box Visualization,Image Contours,Grid Visualization,Image Slicer,Heatmap Visualization,Stability AI Outpainting,Color Visualization,Circle Visualization,Perspective Correction,Image Convert Grayscale,Camera Focus,Absolute Static Crop,Roboflow Visual Search Classifier,Camera Calibration,Crop Visualization,Image Threshold,Icon Visualization,Morphological Transformation,Polygon Visualization,Polygon Zone Visualization,Trace Visualization,Model Comparison Visualization,Relative Static Crop,Camera Focus,QR Code Generator,Background Color Visualization,Stability AI Inpainting,Contrast Equalization,Corner Visualization,SIFT,Background Subtraction,Mask Visualization,Reference Path Visualization,Image Slicer,Contrast Enhancement,Roboflow Visual Search,Text Display,Dynamic Crop,Halo Visualization,Image Blur,Stability AI Image Generation,Ellipse Visualization,Triangle Visualization,Pixelate Visualization,Dot Visualization,Image Preprocessing,Stitch Images,Classification Label Visualization,Polygon Visualization,SIFT Comparison,Line Counter Visualization,Halo Visualization,Depth Estimation,Label Visualization - outputs:
Keypoint Visualization,Twilio SMS/MMS Notification,OPC UA Writer Sink,Keypoint Detection Model,Perception Encoder Embedding Model,Qwen 3.6 API,SAM 3,Llama 3.2 Vision,Distance Measurement,Roboflow Visual Search Classifier,Velocity,Image Threshold,Polygon Zone Visualization,CogVLM,BoT-SORT Tracker,MQTT Writer,Contrast Equalization,Google Gemini,Detections List Roll-Up,GLM-OCR,Reference Path Visualization,Twilio SMS Notification,Google Gemma API,Instance Segmentation Model,PTZ Tracking (ONVIF),Detections Classes Replacement,Detections Merge,Detections Filter,Florence-2 Model,Dynamic Crop,Halo Visualization,Image Blur,LMM,SAM 3 Interactive,Seg Preview,SAM 3,Ellipse Visualization,OpenAI,Florence-2 Model,Byte Tracker,Line Counter,LMM For Classification,Time in Zone,Cache Set,Semantic Segmentation Model,OpenAI,Path Deviation,Event Writer,Instance Segmentation Model,Time in Zone,Google Gemma,Byte Tracker,Slack Notification,Heatmap Visualization,SORT Tracker,Email Notification,Circle Visualization,Perspective Correction,Camera Focus,Byte Tracker,Crop Visualization,Polygon Visualization,Detections Consensus,Line Counter,ByteTrack Tracker,Detections Combine,Stability AI Inpainting,OpenAI-Compatible LLM,SAM 3,Google Gemini,Roboflow Vision Events,Detections Transformation,Multi-Label Classification Model,Size Measurement,Local File Sink,Object Detection Model,SAM3 Video Tracker,Dot Visualization,Line Counter Visualization,Time in Zone,Instance Segmentation Model,Depth Estimation,Stitch OCR Detections,Blur Visualization,Anthropic Claude,Segment Anything 2 Model,Color Visualization,Per-Class Confidence Filter,Stitch OCR Detections,Qwen 3.5 API,Trace Visualization,Icon Visualization,Current Time,Cache Get,Model Comparison Visualization,QR Code Generator,Google Gemini,Detections Stitch,Instance Segmentation Model,Qwen-VL,Mask Visualization,MoonshotAI Kimi,Microsoft SQL Server Sink,Text Display,Stability AI Image Generation,Google Vision OCR,Webhook Sink,OpenAI,Pixelate Visualization,Detection Offset,S3 Sink,Classification Label Visualization,MoonshotAI Kimi,Overlap Filter,Polygon Visualization,Anthropic Claude,Roboflow Dataset Upload,Morphological Transformation,Detection Event Log,Bounding Box Visualization,Stability AI Outpainting,Anthropic Claude,Roboflow Custom Metadata,SAM2 Video Tracker,Roboflow Asset Library Attributes,Morphological Transformation,Detections Stabilizer,Path Deviation,Mask Area Measurement,Background Color Visualization,Model Monitoring Inference Aggregator,Clip Comparison,Corner Visualization,GeoTag Detection,Track Class Lock,Moondream2,Roboflow Visual Search,OpenRouter,Qwen3.5-VL,Llama 3.2 Vision,Overlap Analysis,Email Notification,CLIP Embedding Model,Single-Label Classification Model,Triangle Visualization,OpenAI,Pixel Color Count,Roboflow Dataset Upload,OC-SORT Tracker,Image Preprocessing,SIFT Comparison,YOLO-World Model,Halo Visualization,Label Visualization
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
PP-OCR in version v1 has.
Bindings
-
input
images(image): The image to infer on..
-
output
result(string): String value.predictions(object_detection_prediction): Prediction with detected bounding boxes in form of sv.Detections(...) object.parent_id(parent_id): Identifier of parent for step output.root_parent_id(parent_id): Identifier of parent for step output.prediction_type(prediction_type): String value with type of prediction.
Example JSON definition of step PP-OCR in version v1
{
"name": "<your_step_name_here>",
"type": "roboflow_core/pp_ocr@v1",
"images": "$inputs.image",
"text_detection": "<block_does_not_provide_example>",
"text_recognition": "<block_does_not_provide_example>"
}