EZ
EZ2Conv

AI Object Detector

Detect and identify 80+ object types in images using on-device AI. View results with bounding boxes, labels, and confidence scores. Export annotated images or JSON data -- your photos never leave your browser.

Loading AI Object Detector...

How to Use

  1. Upload or drag and drop an image (JPG, PNG, WebP, or GIF up to 10MB) containing objects you want to identify -- on first use, the AI model (~10MB) downloads and caches
  2. Adjust the minimum confidence threshold: higher values show only high-certainty detections, lower values reveal more objects but may include false positives
  3. Detection runs automatically the moment your image loads -- the AI analyzes it entirely in your browser and identifies people, vehicles, animals, furniture, and 80+ other object types
  4. View results with colored bounding boxes, labels, and confidence percentages overlaid on the image; toggle display options to show or hide boxes, labels, and scores
  5. Download the annotated image with visual overlays, or export detection results as a structured JSON file with object names, confidence scores, and bounding box coordinates

Tips

  • The AI can detect 80+ object types including people, vehicles, animals, and common items
  • Lower the confidence threshold to detect more objects (may include false positives)
  • Toggle display options to customize the result visualization
  • First use may take a few seconds to download the AI model

How On-Device Object Detection Works

What this detector does

The AI Object Detector scans a photo and draws a labeled box around every recognizable thing it finds -- a person, a car, a dog, a laptop, a chair. Under the hood it runs the COCO-SSD model through TensorFlow.js, a neural network trained on the COCO dataset to recognize 80 everyday object classes. Each detection gives you three pieces of information: a class name, a bounding box marking where the object sits, and a confidence score from 0 to 100% telling you how sure the model is.

Detection starts the moment an image loads, and the whole analysis happens on-device -- the model file (~10MB) downloads once and is cached for later visits.

When to reach for it

Use it to auto-tag a folder of photos, sanity-check what a dataset actually contains before training your own model, or count how many people or vehicles appear in a scene. It is handy for accessibility work too -- generating a quick text description of an image -- and for anyone prototyping computer-vision ideas without wiring up a Python environment.

Because it exports structured JSON with coordinates, it also fits into a larger pipeline: run the detector, take the boxes, and feed them into cropping, blurring, or annotation scripts.

A concrete example

Drop in a street photo. The model might return person at 92%, car at 88%, and traffic light at 71%, each wrapped in its own colored box. Raise the minimum confidence to 80% and the traffic light disappears from the results, leaving only the two high-certainty detections. Lower it to 40% and fainter, partially hidden objects start showing up -- along with the occasional wrong guess. Download the annotated image to keep the overlays, or export JSON to get the raw class names, scores, and box coordinates.

Notes and limits

COCO-SSD knows 80 classes, so anything outside that list -- a specific dog breed, a brand logo, readable text -- will not be labeled. Overlapping or very small objects are the hardest cases and often slip below the threshold. The confidence slider is your main control: higher values trade recall for precision, lower values do the reverse. If a photo returns nothing, dropping the threshold usually surfaces the objects the model saw but was not confident about.

Frequently Asked Questions

Every image you run through the detector is free, with no limit on how many photos you analyze or how often. The confidence slider, the box/label/score overlays, the annotated-image download, and the JSON export are all open -- nothing is held back behind a paid tier.
The COCO-SSD model runs through TensorFlow.js on your own machine, so the photo you drop in is analyzed by code in the page, not by a remote service. After the one-time model download, your images and the detection results stay on your device.
No. Load a photo and detection begins on its own -- the boxes, labels, and confidence scores appear without any email, sign-in, or registration step in between.
COCO-SSD recognizes 80 object classes from the COCO dataset: people, vehicles (car, truck, bus, motorcycle, bicycle), animals (dog, cat, bird, horse), furniture, kitchenware, electronics, sports gear, and many common household items. Anything outside those 80 classes -- a specific breed, a logo, readable text -- will not be labeled.
It sets the minimum certainty a detection needs before it shows up. At 70% you see only objects the model is quite sure about; at 30% more objects appear, including faint or partly hidden ones, but false positives creep in. It is the fastest way to trade precision for recall.
If nothing is detected, the objects in your photo may sit below the current confidence threshold -- lower it and run again. Very small, overlapping, or unusual-angle objects are the hardest for the model, and subjects outside its 80 known classes will not register at all.