Age & Gender
Predict age bucket and gender for every face.
What does the age & gender detector do?
The freeimgtool age and gender detector looks at every face in the uploaded photo and predicts an age bucket and a binary gender label. It runs two Caffe CNN models published with the original LearnOpenCV age/gender tutorial. The age model classifies into 8 buckets: (0-2), (4-6), (8-12), (15-20), (25-32), (38-43), (48-53), and (60+). The gender model classifies into Male or Female.
The result is an annotated image with coloured boxes and a label per face, plus a JSON table with confidence scores. Use it for entertainment, classroom demos, or data-pipeline prototypes — not for any decision that affects a real person's life.
How to use it
- Click the upload box and pick a photo (up to 5 MB).
- Click Predict age & gender.
- Read the table of predictions.
- Download the annotated image with face boxes and labels.
Why use this detector
Useful as a classroom demo of computer vision, a quick audience-profile guess, or as a starter pipeline for non-critical analytics. The honest framing is "fun visualisation" — not a life-affecting decision tool.
Who uses it
Students learning CNN classification. Marketers spot-checking audience photos. Hobbyists building photo viewers. Anyone curious how an open-source CNN tags their own selfie.
Limits and ethics
Real talk. These models were trained on a public face dataset that does not represent every demographic equally. Accuracy varies by skin tone, lighting, angle, and age. The gender model is binary, which does not reflect the full range of gender identity. Treat the output as approximate and never use it for hiring, housing, lending, dating, or any decision that affects a real person.
Frequently asked questions
Is the detector free?
Yes. No signup, no premium tier. Display ads support the site.
What is the maximum upload size?
Five megabytes per photo. The tool downscales internally for speed.
Why is the first request slow?
The Caffe age and gender weights (~90 MB combined) are downloaded once and cached on the server. Subsequent requests are fast.
Why is the age so far off?
The model picks one of 8 broad buckets and was trained on limited data. Expect rough estimates only.
Does it work on profile photos?
Frontal faces work best. Profiles, sunglasses, and heavy tilt reduce accuracy or get skipped entirely.
Does it identify who is in the photo?
No. The model only classifies age and gender. It does not store, identify, or match identities.
Is my photo stored?
Uploads are kept only long enough to return the annotated image link and are removed automatically. Privacy policy has the details.
Should I use this for real decisions?
No. It is for fun, learning, and prototypes only.