What is 3D product models?
3D product models are digital representations of physical products as three-dimensional objects, typically in glTF or GLB format, used for AR try-on, web visualization, and interactive ecommerce.
The longer answer
A 3D product model includes geometry (the shape), materials (how surfaces respond to light), textures (color, normal, roughness maps), and metadata (dimensions, units, animations). The standard interchange format for the web is glTF (and its binary container, GLB), which most renderers and AR runtimes can read.
Why it matters
AR try-on, configurators, and immersive product pages all depend on a good 3D model. Without one, you have nothing to render. Production-quality models are the bottleneck for most brands wanting to ship AR; the model pipeline is the project, not the AR widget.
How it works
Models can be authored by a 3D artist, captured via photogrammetry (many photos reconstructed into geometry), or generated by AI image-to-3D services (single-image or multi-image to GLB in minutes). Each path trades cost, speed, and quality. For jewelry, accurate scale, clean topology, and faithful materials matter more than polygon count.
Where this fits in WearRoom
WearRoom uses an AI image-to-3D pipeline so a brand can drop in product photos and get a try-on ready model in under a day. We always override the material to a curated PBR preset for the metal type — geometry comes from the AI; the material reading as metal is on us.
Questions
Photogrammetry or AI image-to-3D?
Photogrammetry is more accurate but slower and requires a controlled capture. AI image-to-3D is faster and works on existing product photos. For jewelry, the AI path with a human QA review is currently the best speed-versus-quality balance.
Why glTF or GLB?
glTF is the JPEG of 3D — an open, compact format supported by every major web AR runtime and renderer. GLB is the single-file binary variant, smaller and easier to ship.