Pretrained object detector

Count cars in a photo.

A pretrained detector that finds every instance of cars in an image and returns each one with a bounding box — count the detections and you have your number. Use the API immediately, no training required, then adapt it to your own data when you need more.

Use this detector Try with your photo

Object detection · counts + bounding boxes Image input

nyckel.com/v1/functions/YOUR_FUNCTION_ID/invoke

cars detected 3

Try the cars detector

Drop in a photo and get the count back. No signup, no setup.

Call the cars detection API

Detection functions run behind your own endpoint — create a free account to set one up in the console, then invoke it with any HTTP client:

curl

curl -X POST "https://www.nyckel.com/v1/functions/YOUR_FUNCTION_ID/invoke" \
  -H "Authorization: Bearer $NYCKEL_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"data": "https://example.com/photo.jpg"}'

Python

import requests

# Get an access token: https://www.nyckel.com/docs/api/overview/authentication/
token = "YOUR_ACCESS_TOKEN"

response = requests.post(
    "https://www.nyckel.com/v1/functions/YOUR_FUNCTION_ID/invoke",
    headers={"Authorization": "Bearer " + token},
    json={"data": "https://example.com/photo.jpg"},
)
result = response.json()
print(len(result[0]["boxes"]))  # the count

Example response

[
  {
    "labelId": "label_...",
    "boxes": [
      { "centerX": 0.42, "centerY": 0.31, "width": 0.18, "height": 0.12, "confidence": 0.93 },
      { "centerX": 0.71, "centerY": 0.55, "width": 0.22, "height": 0.15, "confidence": 0.88 }
    ]
  }
]

One box per cars found — the number of boxes is your count, and each box tells you where in the photo that detection is.

API documentation Authentication Pricing Create a free account

More than a demo: this page is one of thousands of pretrained functions on Nyckel, an ML platform for classification and detection. You can invoke functions by API, review predictions, correct them, collect samples from production traffic, and promote any pretrained function to a private custom model — without changing your integration.

Common questions

What does this detector return?

The detector locates each instance of cars it finds in the photo and returns one detection per instance, each with a bounding box. Counting the detections gives you the number — that's what the widget on this page shows — and the boxes tell you where in the image each one is.

How do I know whether this will work for my application?

Honestly: we can't know in advance — it depends on your data stream and how closely it resembles what this detector has seen. The reliable way to find out is to measure it on your own data: start invoking the detector with real traffic, or upload and annotate a set of images in the console — make sure they look like your production photos, not idealized examples. Nyckel's evaluation metrics then show you exactly how it performs on that data before you rely on it.

What happens when it makes a mistake?

No detector is perfect, so Nyckel is built around the correction loop: invokes can be captured for review, you confirm or correct detections in the console, and corrections become training data. Over time the model adapts to your data distribution — accuracy on your traffic improves with use rather than staying fixed.

What does it cost to try?

Trying the detector on this page is free with no signup. Using the API requires a free account, and the free tier covers your first API calls each month — see nyckel.com/pricing for current limits and paid tiers.

Related detectors

Ready to count cars at scale?

Create a free Nyckel account — you'll get detection functions behind a live API endpoint, and a path to a custom model trained on your own data.

Use this detector Read the docs