Infer snippet world-space AABB from semidense metadata or points.
Preference order: 1) ARIA_POINTS_VOL_MIN / ARIA_POINTS_VOL_MAX (or legacy keys points/vol_min, points/vol_max) 2) Axis-aligned bounds of finite semidense points (ignoring padded/NaN entries)
| Name | Type | Description |
|---|---|---|
| tuple[torch.Tensor, torch.Tensor] | None | Tuple of (min, max) tensors on CPU if finite bounds are available, otherwise None. |
# infer_semidense_bounds { #aria_nbv.data_handling.infer_semidense_bounds }
```python
data_handling.infer_semidense_bounds(efm_dict)
```
Infer snippet world-space AABB from semidense metadata or points.
Preference order:
1) ``ARIA_POINTS_VOL_MIN`` / ``ARIA_POINTS_VOL_MAX`` (or legacy keys ``points/vol_min``, ``points/vol_max``)
2) Axis-aligned bounds of finite semidense points (ignoring padded/NaN entries)
## Returns {.doc-section .doc-section-returns}
| Name | Type | Description |
|--------|---------------------------------------------|--------------------------------------------------------------------------------------------|
| | tuple\[torch.Tensor, torch.Tensor\] \| None | Tuple of ``(min, max)`` tensors on CPU if finite bounds are available, otherwise ``None``. |