Skip to main content
Skip to main content

Vision AI Explorer

An interactive demo of theGoogle Cloud Vision API

Detect text, labels, objects, landmarks, logos, faces and explore visual features powered by AI

How it works

  • 1

    Select an image

    Pick a sample, upload an image, or enter a URL.

  • 2

    Analyze with Cloud Vision AI

    Cloud Vision AI models analyze 11 features.

  • 3

    Explore the results

    Interact with the AI-detected features.

Features list

  • Text detection

    Text detection preview

    Extracts printed and handwritten text from images using optical character recognition.

    Text detection

    Output
    • Full text & language codes
    • Word & paragraph hierarchy
    • Spatial bounding polygons
    Filters and sorting
    • Keyword search
    • Language filtering
    Limits

    Unlimited matches

    Disclaimer

    Varies by resolution and angle.

    How-to Guide

  • Label detection

    Label detection preview

    Identifies broad categories, scenes, objects, and concepts within an image.

    Label detection

    Output
    • Descriptions & categories
    • Confidence scores (0-100%)
    • Topicality scores
    Filters and sorting
    • Keyword filter
    • MID deduplication
    • Sorted by confidence & topicality
    Limits

    Up to 50 matches

    Disclaimer

    Inferred tags, not verified truth.

    How-to Guide

  • Object detection

    Object detection preview

    Detects and localizes multiple physical entities with normalized boundary boxes.

    Object detection

    Output
    • Entity names & categories
    • Normalized vertices (0-1)
    • Confidence scores
    Filters and sorting
    • IoU bounding box deduplication
    • Sorted by confidence (descending)
    Limits

    Up to 50 matches

    Disclaimer

    May group adjacent or occluded items.

    How-to Guide

  • Landmark detection

    Landmark detection preview

    Detects popular natural and human-made structures with geographic coordinates.

    Landmark detection

    Output
    • Entity names & descriptions
    • Geographic coordinates
    • Match confidence scores
    Filters and sorting
    • Interactive satellite map
    • Map marker deduplication
    • Navigation deep links
    Limits

    Varies by cataloged locations

    Disclaimer

    Limited to cataloged structures.

    How-to Guide

  • Logo detection

    Logo detection preview

    Detects popular corporate, commercial, and brand logos within an image.

    Logo detection

    Output
    • Recognized brand names
    • Bounding polygon coordinates
    • Detection confidence ratings
    Filters and sorting
    • Confidence score sorting
    • Knowledge Graph MID deduplication
    Limits

    Up to 50 matches

    Disclaimer

    Limited to cataloged brand marks.

    How-to Guide

  • Face detection

    Face detection preview

    Detects faces, facial landmarks, and emotional likelihoods within an image.

    Face detection

    Output
    • Facial landmarks
    • 3D head pose orientation
    • Emotional likelihood ratings
    Filters and sorting
    • Face selection
    • Emotion intensity progress bars
    • Head pose visualization
    Limits

    Up to 50 matches

    Disclaimer

    Non-biometric; no identity matching.

    How-to Guide

  • Explicit content detection

    Explicit content detection preview

    Detects potentially unsafe content across five safety and moderation categories.

    Explicit content detection

    Output
    • 5 safety categories
    • Likelihood ratings
    • Categorical rankings
    Filters and sorting
    • Severity filtering
    • Pass/flag screening status
    Limits

    Applies to entire image

    Disclaimer

    Probabilistic AI safety estimates.

    How-to Guide

  • Web entities and pages

    Web entities and pages preview

    Searches the web for matching image references, visually similar images, and entities.

    Web entities and pages

    Output
    • Matching web pages
    • Visually similar images
    • Web entities with relevance scores
    Filters and sorting
    • Web Risk domain verification
    • Exact & partial matches
    • Relevance sorting
    Limits

    Depends on public search indices

    Disclaimer

    Links to external public web pages.

    How-to Guide

  • Crop hint detection

    Crop hint detection preview

    Suggests crop hints for specified target aspect ratios to optimize framing.

    Crop hint detection

    Output
    • Crop boundaries
    • Saliency importance fractions
    • Confidence scores
    Filters and sorting
    • Interactive aspect ratio switcher
    • Live preview canvas
    • Saliency sorting
    Limits

    Up to 16 user-specified aspect ratios

    Disclaimer

    Saliency-based framing, not artistic.

    How-to Guide

  • Image properties

    Image properties preview

    Computes general image attributes like dominant colors and pixel coverage.

    Image properties

    Output
    • Dominant RGB colors
    • Pixel coverage percentages
    • Numerical color scores
    Filters and sorting
    • Pixel dominance sorting
    • Luminance sorting
    • WCAG contrast evaluation
    Limits

    Varies by image palette

    Disclaimer

    Visual averages across image matrix.

    How-to Guide

  • API response

    API response preview

    Returns the complete, structured JSON response from the Cloud Vision API.

    API response

    Output
    • Full AnnotateImageResponse
    • Multi-feature data
    • Raw prediction values
    Filters and sorting
    • Feature key selection
    • Searchable syntax tree
    • Payload export
    Limits

    Maximum API response payload

    Disclaimer

    Direct raw Google Vision API output.

    How-to Guide

Vision AI Explorer © 2026