This article answers the most common questions and unexpected behaviors you may run into when using face placement — from "Photo not yet saved" messages, to why faces are different sizes across photos, to why a photo extends past its frame.
"Photo not yet saved — try again"
This message appears when you click Detect Face or replace a photo while the previous save is still in progress. Face detection runs against the saved copy of the layer, so Templified needs the new photo to finish saving before it can detect a face.
- Wait a moment for the save to complete — the editor saves automatically when you drop in a photo or make a change.
- Click Detect Face again. In most cases one retry is all that's needed.
- If the message reappears, try saving the template manually (the save button in the toolbar) and then clicking Detect Face.
You are most likely to see this message if you replace a photo again immediately after the previous replacement — for example, swapping photos quickly in succession. Give each upload a moment to land before detecting.
No face detected
If Templified cannot locate a face in the photo, you will see No face detected — drag the face box on the canvas to place it. Your existing face-target box and max content line settings are not changed.
Common causes and fixes:
- The face is too small in the frame. Use a photo where the subject's face is clearly visible and takes up a reasonable portion of the image — a typical headshot or waist-up sports photo works well.
- The face is obscured or at an extreme angle. Helmets, masks, or profiles that barely show the face can prevent detection. Try a more front-facing photo for the detection step, then swap in the real photo.
- Low image quality. Very dark, blurry, or heavily compressed photos may not have enough detail for the detector. Use the best-quality version of the image available.
If detection keeps failing for your typical photos, you can still position the face-target box manually by dragging it on the canvas after clicking Edit face placement. Detection is a starting point, not a requirement for face placement to work.
Photo renders bigger or smaller than expected
With face placement enabled, the rendered scale of a photo is set entirely by the face math: it is the minimum of two values — the scale needed to fit the detected face into the face-target box, and the scale that keeps the photo's content from crossing the max content line. The photo is not stretched to fill the layer frame.
This means photos from different source crops will naturally render at different sizes, and that is by design:
- A photo taken close-up (the face fills most of the frame) renders smaller overall — the face itself is large relative to the photo, so less scaling is needed to seat it in the face-target box.
- A photo taken from a distance (the face is small in the frame) renders larger — more scaling is needed to bring the face up to the target size.
The result is that faces match in size and position across the whole roster, while the amount of body that shows varies with how each photo was cropped — a knee-up photo will show more of the player than a waist-up photo, even though the faces land at the same height. This is the intended behavior: face placement makes faces consistent, not body framing.
If you need a more uniform amount of body to show across your photos, the most reliable solution is to standardize the source crop at the photography stage.
Photo bleeds past the frame edge
By default, a face-placed photo in Show full photo mode is not clipped to the layer's frame — the entire photo is shown, even if parts of it extend beyond the layer edges. The photo's subject is positioned and sized by the face math; clipping is a separate choice you make.
You have two options to control what shows:
- Show full photo (the default) — the whole photo renders without cropping. The subject's face lands at the face-target box; surrounding content extends in any direction it needs to. Overlapping elements from other layers appear on top based on layer order.
- Crop to box — author a fixed crop rectangle on the canvas using the crop-box overlay. Anything outside that rectangle is hidden. The crop box stays fixed on the canvas; when a new photo with a different face position is dropped in, the face moves to the face-target box and the fixed crop cuts the content as usual.
To switch between these options, select the photo layer and find the Cropping dropdown in the properties panel (visible when the fit mode is Top Fill and face placement is on).
Why is the photo narrower than the layer?
The width of a face-placed photo in the render is determined by the face math — the face scale that aligns the subject's face to the face-target box — not by the width of the layer frame. Depending on the source photo's proportions and the face's position within it, the rendered photo may be narrower than the layer, leaving transparent space on one or both sides.
This is normal and expected. The transparent margins on the sides of a narrow photo are not a problem unless a background or graphic layer behind the slot shows through in an unintended way. If that happens:
- Use Crop to box to define an explicit crop that hides the transparent edges.
- Resize the face-target box to a smaller size — a smaller target face means more of the photo shows around it, which may widen the rendered result.
- Adjust the source photos to use a tighter crop relative to the subject's face.
Editor preview vs. final render
The editor and the final render use the same face placement math, so the position and scale of the face should match between what you see in the editor and what you get in the delivered image. If you notice a difference, the most common causes are:
- The template has unsaved changes. The editor preview updates live as you drag overlays, but the render uses the last saved state of the layer. Save the template before rendering and then compare again.
- The design is using a different photo than the template's sample. The face-target box and max content line are set once on the template but applied to each design's photo, which may have a different face position. The editor in a design shows the real result for that design's photo.
- A crop box is set but not visible in the editor's current zoom. If you set a Crop to box rectangle near the edge of the canvas, zoom out or pan to confirm you can see the full crop overlay.
If the editor and render still disagree after confirming the above, try opening the template, making a small save (such as nudging and nudging back a layer), and rendering again to ensure the latest settings were picked up.
Comments
0 comments
Article is closed for comments.