# OmniPart + TRELLIS2 Assembly Diagnostics

- Run: `/data/250010098/physxanything_stage1_experiments/runs/omnipart_trellis2_perpart_formal_softfit_selected_20260417/000ec112ae7f4a8a93f847ccfd4031be`
- Success parts: `2 / 2`
- OmniPart boxes: `7`

## Part Stats

| idx | part_id | crop | target volume | mask px | faces | mesh/target IoU | fill ratio |
| --- | --- | --- | ---: | ---: | ---: | ---: | ---: |
| 0 | 1 | dim_context | 0.048631 | 19105 | 602438 | 0.6999 | 0.9180 |
| 1 | 2 | dim_context | 0.001456 | 3919 | 212674 | 0.6025 | 0.7295 |

## Pairwise Target BBox Relations

| part_i | part_j | bbox IoU | intersection volume | gap |
| --- | --- | ---: | ---: | ---: |
| 1 | 2 | 0.0299 | 0.001456 | 0.000000 |

## Notes

- These diagnostics are AABB-level sanity checks, not surface-intersection tests.
- High target bbox IoU means assembly conditions overlap before any mesh generation.
- Low mesh-to-target bbox IoU or very anisotropic extent ratios can indicate poor bbox fit or part hallucination.