Incoming Transformations
transformation: [...]Resize, crop, and optimise the moment the file lands — before it is stored. No per-request cost.
TanStack Start · Cloudinary
Build a community photo wall — the right way.
Users upload anything. Your app resizes it, watermarks it, and only shows approved photos. Cloudinary handles the hard part — you handle the experience.
The problem
Managing user-generated content is scary. Users upload huge files, wrong formats, or portraits that crop terribly. You need to resize, watermark, and moderate — before anything hits your storage.
The solution
Cloudinary incoming transformations fire the moment a file arrives — not at request time. The stored image is already resized, watermarked, and face-cropped.
Cloudinary powers every upload
Four features, zero effort on your part
Face Detection Crop
g_auto:facesGravity set to auto:faces so the crop always centres on the most prominent detected face.
Logo Watermark
l_logo / text overlayYour brand stamp is baked into the stored file at upload time — not rendered at delivery.
Auto Format & Quality
f_auto, q_autoServe WebP or AVIF based on browser support. Cloudinary picks the smallest file with good fidelity.
Incoming transformations
Applied once, at upload time
Unlike eager or on-the-fly transforms, incoming transformations run before the file is stored. The result is baked in — delivery is instant and there are no per-request compute costs.
📁User file
→🛡️Client check
→type + size☁Cloudinary
→upload_stream↔Resize
→c_fill, w_800👤Face crop
→g_auto:faces💧Watermark
→l_logo⚡Optimise
→f_auto, q_auto🗄️Stored
pending⏳pending
Awaiting review
Every upload starts here. The photo is stored and processed but not yet public.
✅approved
Visible in gallery
Moderator approved it. The photo appears on the public wall for everyone to see.
🚫rejected
Hidden from gallery
Moderator rejected it. The photo stays stored but never appears publicly.