FAQ

Does my photo get uploaded to your server?

No. Your photograph never leaves your device. Every face tool on The Face Shape Detector runs its analysis inside your own browser using Google MediaPipe, a landmark model the browser downloads once and then executes locally. There is no photo upload endpoint on our servers, so no address exists for an image to be sent to, even by mistake. You can confirm it yourself in about a minute: open your browser's developer tools, switch to the Network tab, and run an analysis. The model file downloads. No image goes anywhere.

Where exactly is the processing done?

Processing happens on the processor inside the device you are holding. When you pick a photo, the browser reads the file into memory, draws it to an in-page canvas, and hands those pixels to the MediaPipe model already loaded in the tab. The model returns 468 coordinate points, and The Face Shape Detector's code computes ratios from them. All of it stays inside the browser's own sandbox. No step in the chain sends a network request carrying your image, and closing the tab discards everything.

Do you store anything about me at all?

Nothing from the tools. Photo analysis and quiz answers alike stay on your device, so color analysis, body shape, body type and Kibbe results are never written anywhere we can read. What The Face Shape Detector does collect is narrow and ordinary: contact form submissions, saved to a database so we can reply; standard server access logs holding IP address, timestamp and browser type; Google Analytics measurement data; and advertising data collected by Google AdSense. The public site has no account system and no login, so no profile of you exists to build.

Do you use my photo to train models?

No, and it would be impossible. Training needs collected images, and none ever reaches us. The landmark model behind the face tools is Google's pre-trained MediaPipe Face Mesh, which arrives at your browser already trained, stays unmodified by anything you do, and sends nothing back. No image analyzed here is stored, logged, sampled, queued for review or added to a dataset, ours or anyone else's. That follows from the architecture rather than from a promise somebody has to police.

Is the camera option any different?

No, the camera path works the same way. Grant camera permission and your browser hands the video stream to the page, which renders it into a local canvas frame by frame for the model to read. Frames are analyzed and discarded immediately. Nothing is recorded, and nothing is transmitted. Most browsers scope that permission per site and per session, and you can revoke it at any moment from the padlock or camera icon in the address bar. Prefer not to grant it? Use the file picker instead, for an identical result.

Is it safe to use on a shared or public computer?

Broadly yes, with one caveat about the machine rather than the site. Because no image is uploaded, nothing about your face persists on our side. What can persist locally is browser state: a theme preference in local storage, quiz progress in session storage, and analytics or advertising cookies. And if you picked a photo from the machine's own file system, that file is still sitting there. Use a private or incognito window, close the tab when you finish, and delete any file you copied onto the device.

What are the 468 landmarks?

The 468 landmarks are numbered points that the MediaPipe Face Mesh model places on a detected face: along the jawline, around each eye, across the brows, down the nose bridge, around the lips and over the cheeks. Each carries an x, a y and an approximate z coordinate. Once the mesh is fitted, geometry becomes arithmetic. Face length is the distance between two points, jaw width the distance between two others, and a face shape classification is a set of ratios between those distances compared against defined ranges.

How accurate is it?

Landmark placement on a clear, front-facing, evenly lit photograph is reliable to within a few pixels, and the arithmetic that follows is exact. The uncertainty sits elsewhere. Face shape categories have no agreed scientific boundary, so where one shape ends and the next begins is a judgment encoded in our thresholds rather than a fact about faces. Expect a dependable answer when your proportions sit clearly inside one category, and genuine uncertainty when they sit near a boundary. That is why confidence percentages get published alongside every result.

Why did I get a different result from a different photo?

Because the geometry the model measured genuinely changed between the two shots. A head tilt of a few degrees alters width-to-height ratios. Side lighting darkens one edge of the jaw and moves the outline. A smile widens the lower face. A close-range selfie lens enlarges central features relative to the perimeter. Hair across the forehead hides a measurement the model needs. When two photographs disagree, take a third: square to the camera, neutral expression, even front light, hair back, camera at arm's length or further.

Why does it show percentages instead of one answer?

Because a single answer would misrepresent the certainty. Most faces are not textbook examples of one category; they sit between two. A result of 44 percent oval and 38 percent heart is not the tool failing to decide. It is an accurate description of a face with oval proportions and a heart-shaped hairline. Reporting only "oval" would throw away the more useful half of the finding. Read the top two figures together, and where they sit close, treat both sets of styling guidance as relevant to you.

Do I need to install anything?

No. There is no app, no plugin, no extension and no account. Everything runs in a standard web browser on a page you have already loaded. The only thing that downloads is the MediaPipe model file, which your browser fetches automatically and caches like any other web asset. Camera access is optional too: you need to grant it only if you choose the camera option rather than picking a file from your device. Nothing is charged for any of this, at any point.

Does it work offline?

Partly, and not by design. Your first visit needs a connection to load the page and the model file. Once your browser has cached that model, an analysis may still run while you are offline, because the computation itself needs no network at all. We do not guarantee it. Cache behavior varies between browsers, and a cached model can be evicted at any time. Treat offline operation as an occasional side effect rather than a feature you can plan around.

What makes a good photo?

A good photo faces the camera square on, with your eyes level and a neutral, relaxed expression. Use even light from the front. A window during daylight works well, while overhead spotlights and strong side light cast shadows the model may read as an edge. Pull hair away from your forehead and jaw. Hold the camera at arm's length or further to avoid lens distortion. Aim for at least 600 pixels across your face, unfiltered and uncropped. Those five conditions account for most of the variation between two results.

Does it work with glasses, a beard or a headscarf?

Yes, with a caveat for each. Glasses bias the eye landmarks, so remove them for the eye shape tool and keep them on when you are testing frame styles. A beard changes the visible jaw outline, so a face shape result describes the bearded outline rather than the bone beneath it, which is useful for choosing a haircut and less useful for reading your structure. A headscarf covering the hairline hides the forehead measurement, lowering confidence in face shape but not in eye, nose or lip analysis.

Which browsers and devices are supported?

Current versions of Chrome, Edge, Safari and Firefox on Windows, macOS, Android, iOS and iPadOS. The tools rely on WebAssembly and, where available, WebGL acceleration, both standard in browsers released over the last several years. Older devices work, just more slowly. Ancient browsers, some heavily locked-down corporate networks, and privacy modes that block WebAssembly can stop the model loading at all. When a tool refuses to start, try a different browser before reporting it, because that resolves most cases.

Why is the camera not working?

Permissions, usually. Check the padlock or camera icon in your address bar, confirm the site is allowed to use the camera, then reload the page. On iOS, camera access requires Safari or a browser built on Safari's engine, and it fails inside some in-app browsers such as those in social media apps, so open the page in Safari or Chrome directly. Also confirm that no other application is holding the camera, and that your operating system's privacy settings permit browser camera access at all.

Why is the first analysis slower than the rest?

Because the first run downloads the landmark model, initializes it, and compiles it for your device. That model is several megabytes, so expect a one-off cost of a few seconds on a good connection and longer on a slow one or an older phone. Later analyses reuse the model already sitting in memory and typically finish in well under a second. If it feels slow again on a return visit, your browser has evicted the cached model and is fetching it once more.

What if I am between two shapes?

Then you are between two shapes, which is a legitimate outcome rather than an error. Most people are. When the top two confidences land within roughly ten points of each other, read both descriptions and both sets of styling guidance, then take from each whatever matches what you see in the mirror. In practice the recommendations for adjacent shapes overlap heavily, so a narrow margin rarely changes which haircut, frame or neckline suits you.

Is one face shape better than another?

No. Oval gets described as the most versatile shape in styling literature, because a wide range of haircuts and frames suit it without adjustment, and that has been misread for decades as "best". Versatility is not superiority. Every shape has proportions that flatter it, and some of the most distinctive faces are the ones furthest from an average. The Face Shape Detector describes geometry so you can make better decisions about hair, frames and makeup. Nothing on the site ranks faces against each other.

Does face shape change with age or weight?

The underlying bone structure is largely settled by early adulthood and shifts little afterwards. Everything around it does change. Weight gain and loss alter cheek and jaw fullness, which can move a measured result between round and oval, or between square and oblong. Aging brings a gradual loss of facial fat and skin elasticity that narrows the mid-face and softens the jawline. So a measurement taken ten years apart can differ even though the skeleton underneath has not moved.

Can I get results for a child?

The tools will run on any detectable face, but we advise against it and the output means little. Classification thresholds derive from adult proportions, and a child's facial proportions change substantially through growth: the forehead is proportionally larger and the jaw far less developed. A result for a child says nothing about the face they will have as an adult. If you do run one, it must be your own child and you must be the person holding the device.

What should I do with the result?

Use it as a starting point for practical decisions: which glasses frames to try on, whether a fringe will suit you, where a stylist might add or remove volume, which necklines balance your proportions. Take the confidence figures seriously, because a narrow margin means keeping your options open. And treat the whole thing as one input among several, alongside your own eye and, where it matters, a professional's. It describes geometry. It is not an instruction.

How do the quiz tools differ from the photo tools?

Quiz tools use no camera and no image at all. The color analysis quiz, body shape calculator, body type calculator and Kibbe body types tool ask you questions or take measurements you enter, then apply a rule set to those answers. Everything is computed in your browser and nothing is stored or sent to us. The trade-off differs too: photo tools measure objectively but can be thrown by lighting and angle, while quiz tools depend entirely on how accurately you observe and measure yourself.

How accurate is self-measurement?

Self-measurement is the main source of error in these tools, and it is easy to improve. Measure over close-fitting clothing rather than a jumper. Keep the tape level and snug without compressing. Take each measurement twice. For shoulders, measure the widest point across the back; for waist, use the narrowest point rather than where a waistband happens to sit. Small errors matter because these frameworks turn on ratios, and a two-centimeter mistake can push a borderline result into the next category.

Are Kibbe types and color seasons scientific?

No, and The Face Shape Detector does not present them that way. Kibbe body types come from a 1987 style book by David Kibbe. Seasonal color analysis derives from mid-century color theory popularized in the 1980s. Somatotypes originate in W. H. Sheldon's 1940s work, whose claims linking build to temperament are discredited and are not repeated here. These are descriptive styling vocabularies, genuinely useful for discussing proportion, contrast and line, but they hold no clinical or biological standing.

Is the site free?

Yes, entirely. All 14 tools are free, with no account, no trial, no usage limit and no paywall on any result. Nobody is asked to register, to hand over an email address to see an outcome, or to pay for a fuller version of an analysis. No premium tier exists. Everything the site can tell you appears on the page the moment the analysis finishes, and the same is true for every article published here.

Why are there ads?

Because advertising pays for the site. Hosting, the domain, the writing and the development time all cost money, and Google AdSense covers that without charging you or harvesting your data. It is a deliberate trade: ads let the analysis stay client-side and the tools stay free, rather than forcing an account system and a subscription. Advertising never influences an editorial conclusion, and no payment buys coverage inside an article or a favorable mention in a tool recommendation.

Can I embed a tool on my own site?

Often yes. Salons, opticians, stylists and educators ask regularly, and because the analysis is client-side, nothing has to be proxied through your server. Write to us with the tool you want, the page it would sit on and that page's rough monthly traffic, and we will confirm terms in writing. We ask for attribution and a link to the source page, and we ask that results are never presented in a way implying a professional or diagnostic assessment.

Can I use my results commercially?

Your own result is yours. Put it in a video, a blog post or a client consultation as you like, with attribution to The Face Shape Detector. What you may not do is scrape the tools, wrap them inside another product, or resell access to them. If you want to build a commercial workflow around the analysis, that is an embedding conversation rather than a scraping one, and we would much rather have it in advance than discover it afterwards.

How do I report a problem?

Use the contact form and include the page URL, what you expected, what happened instead, and your browser, version, device and operating system. For photo tools, describe the conditions (lighting, angle, glasses, hair) rather than sending an image, because our systems cannot receive photographs and we do not want them by email. Accessibility barriers take priority over other bug reports and get acknowledged within two business days, with a plan and a date attached.

How do I contact you?

Through the contact page, or by writing directly to [email protected]. Both reach the same inbox, and a person reads it. Expect a reply within two to three business days, with weekends and holiday periods running longer. Please keep medical questions out of it, because we are not clinicians and cannot answer them. And please do not attach photographs, which are deleted unopened.

Six themes, recapped. The Face Shape Detector runs 14 free browser-based tools, the face ones built on Google MediaPipe Face Mesh and its 468 numbered points, the quizzes built on answers you type. Privacy comes first because the answer is architectural: your photograph never leaves your device, no upload endpoint exists, and nothing you analyze is stored, logged or used for training.

Accuracy is exact in the arithmetic and uncertain at the category boundaries, which is why percentages appear instead of a single verdict. Photograph conditions, glasses, beards, browsers and camera permissions all shape what you get. Landing between two shapes is normal, no shape ranks above another, and results for children mean little. Quizzes stand or fall on your own tape measure. Everything is free, funded by Google AdSense, and embeddable on request.

Still stuck? The contact form and [email protected] both reach a person.