Free Face Shape Detector Guide
A face shape detector is most useful when it explains why a photo reads as oval, round, square, heart, diamond, oblong, or triangle instead of only giving a label.
Updated August 17, 2026. iLook results describe one submitted photo and are not medical or biometric measurements.
What this free tool reads
- The visible outline from forehead to cheekbones, jawline, and chin.
- The closest face-shape category plus the next likely alternative.
- How camera angle, hair, beard, crop, and expression can change the outline.
How to interpret the output
Use the result as styling context. Face shape does not change the overall iLook score, and one photo should not be treated as a fixed identity label.
The strongest use case is comparison against your own photos: better light, cleaner framing, less tilt, and a clearer crop. A single score should never be treated as a judgment of a person.
Best photo setup
Use one clear adult face, neutral light, a front-facing pose, and a crop that includes the forehead, eyes, cheeks, jaw, and chin. Avoid filters, heavy shadows, extreme angles, face-obscuring accessories, and multiple people in the frame.
Privacy and responsible use
Upload only authorized adult photos. iLook rejects unclear, multiple-face, possible-minor, unsupported, or low-quality submissions.
Successful website and app scans stay in the submitter's private history until deletion, while API-key requests remain transient. iLook does not identify the person, infer protected traits, or create face embeddings or a landmark archive.
Try the free tool
Run the focused scanner here: Face shape analyzer. You can also use the main iLook face analyzer for the full report with all free-tool breakdowns.
What is the best photo for a face shape detector?
Use a centered, front-facing adult photo with the forehead, cheeks, and jaw visible. Avoid heavy tilt, sunglasses, hats, and shadows.
Can my face shape result change between photos?
Yes. Lens distance, expression, crop, hairstyle, beard shape, and head angle can make the visible outline read differently.