Skip to Content
Framing WorkbenchFrame Image Optimization

Frame Image Optimization

Real-world moulding photos often suffer perspective distortion and uneven lighting, which makes renderings look fake. Frame image optimization is an image pipeline that turns one photo into a tileable rendering asset.

Where to find it

The “Optimize” button in the frame/fillet add & edit forms of the asset library.

Flow

  1. Clicking “Optimize” starts a task; the progress bar shows Queued → Detecting → Perspective correction → Done (polled every 0.8s, ~2 min cap);
  2. High detection confidence (≥ 0.8) continues automatically; otherwise a “Need crop” step appears — drag the four corners / edges / whole box to frame the strip, then “Confirm crop” (or “Skip crop”);
  3. When done you get a “Framing preview” and a “Before / after” comparison;
  4. Click “Use optimized result” to replace the original, or “Keep original” to discard.

Notes

  • The source must be an image uploaded through the platform, longest edge ≤ 8000px;
  • After adopting, a “Restore original” link appears in the form and applies on save; uploading a new image clears the original record;
  • Closing the dialog without adopting cleans up temporary files automatically;
  • Concurrency is limited (up to 3 running tasks per user); tasks expire after 1 hour.