Skip to content

Blur Visualization

Class: BlurVisualizationBlockV1

Source: inference.core.workflows.core_steps.visualizations.blur.v1.BlurVisualizationBlockV1

Apply blur effects to detected objects in an image, obscuring their details while preserving the background, useful for privacy protection, content filtering, or visual emphasis.

How This Block Works

This block takes an image and detection predictions and applies a blur effect to the detected objects, leaving the background unchanged. The block:

  1. Takes an image and predictions as input
  2. Identifies detected regions from bounding boxes or segmentation masks
  3. Applies a blur effect (using average pooling) to the detected object regions
  4. Preserves the background and areas outside detected objects unchanged
  5. Returns an annotated image where detected objects are blurred, while the rest of the image remains sharp

The block works with both object detection predictions (using bounding boxes) and instance segmentation predictions (using masks). When masks are available, it blurs the exact shape of detected objects; otherwise, it blurs rectangular bounding box regions. The blur intensity is controlled by the kernel size parameter, where larger kernel sizes create stronger blur effects. This creates a visual effect that obscures or anonymizes detected objects while maintaining context from the surrounding image, making it ideal for privacy protection, content filtering, or focusing attention on the background.

Common Use Cases

  • Privacy Protection and Anonymization: Blur faces, people, license plates, or other sensitive information in images or videos to protect privacy, comply with data protection regulations, or anonymize content before sharing or publishing
  • Content Filtering and Moderation: Obscure inappropriate or sensitive content in images or videos for content moderation workflows, safe content previews, or user-generated content filtering
  • Visual Emphasis and Focus: Blur detected objects to draw attention to other parts of the image, create visual contrast between blurred foreground objects and sharp backgrounds, or emphasize specific elements in composition
  • Product Photography and E-commerce: Blur detected distracting elements or secondary products in images to keep the main subject sharp and prominent for product photography, catalog creation, or e-commerce image preparation
  • Security and Surveillance: Anonymize people, vehicles, or other identifiable elements in security footage or surveillance images while preserving scene context for analysis, reporting, or public sharing
  • Documentation and Reporting: Create anonymized or censored versions of images for reports, documentation, or case studies where sensitive information needs to be obscured but overall context should remain visible

Connecting to Other Blocks

The annotated image from this block can be connected to:

  • Other visualization blocks (e.g., Label Visualization, Bounding Box Visualization, Polygon Visualization) to add additional annotations on top of blurred objects for comprehensive visualization or to indicate what was blurred
  • Data storage blocks (e.g., Local File Sink, CSV Formatter, Roboflow Dataset Upload) to save blurred images for documentation, reporting, or archiving privacy-protected content
  • Webhook blocks to send blurred images to external systems, APIs, or web applications for content moderation, privacy-compliant sharing, or anonymized analysis
  • Notification blocks (e.g., Email Notification, Slack Notification) to send blurred images as privacy-protected visual evidence in alerts or reports
  • Video output blocks to create annotated video streams or recordings with blurred objects for live monitoring, privacy-compliant video processing, or post-processing analysis

Type identifier

Use the following identifier in step "type" field: roboflow_core/blur_visualization@v1to add the block as as step in your workflow.

Properties

Name Type Description Refs
name str Enter a unique identifier for this step.. ❌
copy_image bool Enable this option to create a copy of the input image for visualization, preserving the original. Use this when stacking multiple visualizations.. ✅
kernel_size int Size of the blur kernel used for average pooling. Larger values create stronger blur effects, making objects more obscured. Smaller values create subtle blur effects. Typical values range from 5 (light blur) to 51 (strong blur). Must be an odd number for optimal blurring performance.. ✅

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 Blur Visualization in version v1.

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Blur Visualization in version v1 has.

Bindings

  • input

  • output

    • image (image): Image in workflows.
Example JSON definition of step Blur Visualization in version v1 
{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/blur_visualization@v1",
    "image": "$inputs.image",
    "copy_image": true,
    "predictions": "$steps.object_detection_model.predictions",
    "kernel_size": 15
}