Choose the asset representation that best serves gameplay, readability, iteration speed, and runtime budget. Do not force every reference image through image-to-3D reconstruction.
Classify the asset before making it
| Need | Preferred representation |
|---|---|
| Hero character or complex authored animation | Imported rigged mesh, with verified license and scale |
| Weapon, relic, pickup, creature variant, architecture, or interaction prop | Procedural Three.js geometry when controllable silhouette, sockets, collision, or code-driven variants matter |
| Concept, class portrait, inventory icon, cursor, backdrop, or menu treatment | Generated or authored 2D image/video used directly as UI media |
| Level mood, composition, or spacing reference | Concept image used as visual guidance; author the actual playable level separately |
| Reference image whose visible form itself must become a code model | Use img2threejs with its quality gates and state the result is a reconstruction, not extracted geometry |
Create image-first work accurately
Generate a compact reference set: hero view, silhouette/side view when geometry matters, palette/material close-up, and usage context. Define camera distance, scale, pivot, collision, attachment sockets, animation needs, and budget before implementation. A single image cannot establish hidden geometry; request more views or select deliberate stylization.
Build the runtime asset
- For imported meshes, normalize coordinate system, scale, pivot, materials, animation clips, and ownership/license before scene integration.
- For procedural geometry, expose a factory with named parts, stable sockets, deterministic variation, collider guidance, and reusable materials.
- For 2D assets, export transparent media at the actual display density; provide alt text/labels where user-facing and avoid using raster art as fake 3D collision.
- Keep authored content identifiers separate from runtime state. Do not duplicate item, material, or gameplay facts in UI code.
Verify in the game, not the asset viewer
Check the real camera distance, lighting, readability, collision/contact, animation, mobile viewport, and performance. Measure representative draw calls, triangles, texture count, and frame time. Keep visuals original: use references for visual grammar, then create new identity, geometry, and media.
Record provenance
Track whether each runtime asset is imported, procedural, generated 2D, or reference-only, together with source/license and generation input where applicable. Do not claim an image-to-model pipeline unless source and runtime code actually demonstrate one.