Skip to content

Bounds

Purpose

Owns scanning an animated prefab's clips to find the largest on-screen pose.

This territory samples poses across the discovered animation clips and measures each pose's true deformed extent (via SkinnedMeshRenderer baking), returning the world-space envelope that contains every sampled pose, without owning animation sampling, camera framing, or export.

Owns

  • Per-pose deformed bounds measurement (skinned-mesh baking)
  • Largest-pose / pose-envelope scanning across clips
  • Aggregation of sampled poses into a single fit envelope

Does NOT Own

  • Animation sampling ownership (see Animation/Sampling)
  • Camera framing / fitting (see Core/Utilities, pipeline camera managers)
  • Export orchestration or capture execution
  • Panel UI layout

Allowed Dependencies

Bounds may depend on:

  • Unity rendering APIs (Renderer, SkinnedMeshRenderer.BakeMesh, Bounds)
  • The animation sampling utility (to pose the rig before measuring)
  • Lightweight bounds/result types

Bounds should avoid depending on:

  • export orchestration systems
  • camera ownership systems
  • panel/UI systems

Notes

This folder owns measuring the largest pose, not acting on it.

Camera framing systems decide: - how the camera is positioned to the resulting envelope

Panels decide: - when a scan runs and how the result is surfaced

Bounds own: - accurate per-pose extent measurement (baked, not bind-pose) - the envelope that contains every sampled pose

Future Candidates

  • Keep this territory measurement-only; do not absorb camera or export ownership.
  • Consider caching scan results if repeated scans become a performance concern.
  • Maintain bake-based accuracy so animated extents are measured, not bind-pose bounds.