Buckets:
| # Scroll geometry: lasagna fields, spiral fit surfaces, m7 surface predictions and dinovol slabs | |
| Geometry computed from public Vesuvius Challenge CT scans with the community's open models and code, so you can | |
| reuse it without running a GPU job yourself: one villa lasagna field, villa spiral fit surfaces, m7 surface | |
| predictions and two dinovol embedding slabs. Everything here is **Model output, not a reading**. No file holds ink | |
| labels, ink maps, hand-made annotations or CT voxels, and no surface here has been checked by hand. | |
| All scans are Vesuvius Challenge open data. None of the scrolls here is on the prize eligibility list | |
| (Sourced fact: [villa `prizeEligibility.json` at `deb1b695`](https://github.com/ScrollPrize/villa/blob/deb1b6953d6725966c57bfe40feddbcd14644858/scrollprize.org/src/data/prizeEligibility.json), read 3 Nov 2026, and rechecked on villa `main`, 11 Nov 2026). | |
| ## What is in it | |
| Paths are `KIND/SCROLL/VOLUME/...`. `VOLUME` is the open-data volume id; `-L2` means level 2 of that volume. | |
| | Folder | Scan and volume | Region (voxels of that level) | What | Size | | |
| | --- | --- | --- | --- | --- | | |
| | `lasagna-fields/PHerc0172/20241024131838/z9024-10560/` | PHerc0172, `20241024131838-7.910um-53keV-masked.zarr`, level 0 | z 9,024 to 10,560, whole slice | villa lasagna `predict3d` field in 5 parts | 9.15 GB | | |
| | `spiral-fits/PHerc0332/20251211183505-L2/z3200-3712/` | PHerc0332, `20251211183505-2.399um-0.2m-78keV-masked.zarr`, level 2 (9.596 um) | z 3,200 to 3,712 | 9 spiral fit surfaces (seeds 1 to 3, variants A, A-repeat, AV) | 0.44 GB | | |
| | `spiral-fits/PHerc0332/20251211183505-L2/z2432-3456/` | same volume, level 2 | z 2,432 to 3,456 | 3 spiral fit surfaces (seeds 1 to 3, variant A) | 0.34 GB | | |
| | `spiral-fits/PHerc0172/20241024131838/z3004-4028/` | PHerc0172, `20241024131838-7.910um-53keV-masked.zarr`, level 0 | z 3,004 to 4,028 | 3 spiral fit surfaces (seeds 1 to 3, variant A) | 0.32 GB | | |
| | `spiral-fits/PHerc0172/20241024131838/z9280-10304/` | same volume, level 0 | z 9,280 to 10,304 | 18 spiral fit surfaces (seeds 1 to 3, variants A, E, G, Q, G-pilot, Q-pilot) | 3.36 GB | | |
| | `m7-surface/PHerc0009B/20260319104112-L2/crops/` | PHerc0009B, `20260319104112-2.401um-0.3m-77keV-masked.zarr`, level 2 | 48 boxes of 384 voxels | m7 surface masks, plus the unthresholded probability of one 256-voxel cube per box | 1.29 GB | | |
| | `m7-surface/PHerc0332/20251211183505-L2/crops/` | PHerc0332, `20251211183505-2.399um-0.2m-78keV-masked.zarr`, level 2 | 48 boxes of 384 voxels | as above | 1.38 GB | | |
| | `m7-surface/PHerc0841/20260319124803-L2/crops/` | PHerc0841, `20260319124803-2.403um-0.2m-77keV-masked.zarr`, level 2 | 48 boxes of 384 voxels | as above | 1.74 GB | | |
| | `m7-surface/PHerc0841/20250821151531/crops/` | PHerc0841, `20250821151531-9.366um-1.2m-113keV-masked.zarr`, level 0 | 48 boxes of 384 voxels | as above | 1.80 GB | | |
| | `m7-surface/PHercMAN5/20260311104824-L2/crops/` | PHercMAN5, `20260311104824-2.399um-0.2m-78keV-masked.zarr`, level 2 | 48 boxes of 384 voxels | as above | 1.43 GB | | |
| | `m7-surface/PHerc0172/20241024131838/z9600-9984/` | PHerc0172, `20241024131838-7.910um-53keV-masked.zarr`, level 0 | z 9,600 to 9,984, 67 boxes | m7 surface masks | 2.33 GB | | |
| | `m7-surface/PHerc0172/20241024131839/z12864-13824/` | PHerc0172, `20241024131839-7.910um-53keV-masked.zarr`, level 0 | z 12,864 to 13,824, y 768 to 2,304, x 3,264 to 5,568 | m7 surface mask, one box | 0.37 GB | | |
| | `m7-surface/PHerc1667/20231117161658/z4032-5184/` | PHerc1667, `20231117161658-7.910um-53keV-masked.zarr`, level 0 | z 4,032 to 5,184, 8 boxes | m7 surface masks | 0.16 GB | | |
| | `dinovol-embeddings/PHerc0009B/20250521125136-L0/z3360-3552.zarr` | PHerc0009B, `20250521125136-8.640um-1.2m-116keV-masked.zarr`, level 0 | z 3,360 to 3,552, whole slice | dinovol `dense` embeddings | 0.52 GB | | |
| | `dinovol-embeddings/PHercMANBp/20251216152116-L2/z1568-1760.zarr` | PHercMANBp, `20251216152116-2.399um-0.2m-78keV-masked.zarr`, level 2 | z 1,568 to 1,760, whole slice | dinovol `dense` embeddings | 0.07 GB | | |
| The volumes are at `https://vesuvius-challenge-open-data.s3.us-east-1.amazonaws.com/SCROLL/volumes/` (PHerc1667's | |
| `20231117161658` volume is at `https://data.aws.ash2txt.org/samples/PHerc1667/volumes/`). | |
| ## The files | |
| **Lasagna field.** `tNNN/PHerc0172-lasagna-tNNN.tar` holds one part of the field as written by villa's | |
| `lasagna-preprocess predict3d` (OME-Zarr groups `PHerc0172_cos`, `_grad_mag`, `_nx`, `_ny` and the manifest | |
| `PHerc0172.lasagna.json`); `tNNN/meta/` is a copy of each part's small metadata files. To use the field, unpack all | |
| five tars into one folder, then copy `PHerc0172.lasagna.json` from the field's folder over the one the tars wrote: it | |
| is the joined manifest that lists all five crops. Only z 9,024 to 10,560 was predicted; a few edge chunks of parts | |
| t003 and t004 reach past z 10,560. The parts' sha256: | |
| | Part | Crop z, y, x | sha256 | | |
| | --- | --- | --- | | |
| | t000 | 9,024 to 9,216, 0 to 6,700, 0 to 9,100 | `755e6aa86f9f102036fc4c935805f2fcab213a9161b5a2bd9c7566e0e35b8c4f` | | |
| | t001 | 9,216 to 10,240, 0 to 6,700, 0 to 4,096 | `8059c8303332175bd8c1f9da8c04a4871baf16983a19f5eade6813ad4aefc995` | | |
| | t002 | 9,216 to 10,240, 0 to 6,700, 4,096 to 9,100 | `4aa087d11e265f5952bdb5d66dd60eb6da18d0c5f4ad1fb64231ff1d67e3fcce` | | |
| | t003 | 10,240 to 10,560, 0 to 6,700, 0 to 4,096 | `b25e8aea57d3ccd587c644d7259e08f1cfff00c8cbed6ad9084f2be69227e7da` | | |
| | t004 | 10,240 to 10,560, 0 to 6,700, 4,096 to 9,100 | `09325b1d552d20a2daf42781aa74b05558552e5f335f7cc863405be693853cd4` | | |
| **Spiral fit surfaces.** Each `seedN-VARIANT/SCROLL_zA-B_sN.tifxyz/spiral-checkpoint.tifxyz/` is one tifxyz surface | |
| (`x.tif`, `y.tif`, `z.tif` and `meta.json`), as written by villa's spiral fit export. Some coordinate files are | |
| stored zstd compressed as `x.tif.zst` (and `y`, `z`); `zstd -d` gives the original TIFF. Coordinates are voxels of | |
| the volume and level in the folder name. The variants: | |
| - **A**: villa's spiral fit with its default settings, 30,000 steps, on a villa lasagna field of the same scan. | |
| - **A-repeat**: A again with the same seed, to show run to run spread. | |
| - **AV**: villa's spiral fit with flow-gradient smoothing alone. | |
| - **E, G and Q**: villa's spiral fit given relative windings (villa's `relative_windings.json`) made by the rule of | |
| Ringleader 1.1.0's `ladders.py` ([page](https://github.com/ChaseHendrick/Scrolls/blob/main/docs/tools/ringleader.md)): | |
| E with round 1 ladders, G with round 2 grown ladders, and Q with G's plus-one control, which should make the fit | |
| worse. The ladder files are not included. | |
| - **G-pilot and Q-pilot**: a second, separate run of G and Q with the same seeds. | |
| They are here as surfaces to reuse or compare. This card makes no claim that one variant is better. | |
| **m7 surface predictions.** Each box folder holds `pred.zarr`, an OME-Zarr (zarr v2) group with the full shape of | |
| the scan level and levels 0 to 5, where only the box's 192-voxel chunks plus a halo of one chunk on each side are | |
| stored; values are 0 or 255 (m7's foreground probability at threshold 0.2). Folders named `crops/` also hold | |
| `tNN/prob.npz` with `prob` (uint8, 256 x 256 x 256, m7's foreground probability times 255, no threshold) and `origin` | |
| (the cube's z, y, x origin). `tiles.json` in each folder lists every box, the cube origins and the run settings. | |
| **dinovol slabs.** zarr v2 groups with `dense` (int8, z x y x x x 64) and `dense_scale` (float16, z x y x x): | |
| `value = dense * dense_scale`, then each vector has unit length. Index `i` along an axis covers voxels | |
| `[origin + 4 i, origin + 4 i + 4)` of the level named in the folder, with origin z 3,360 (PHerc0009B) or 1,568 | |
| (PHercMANBp) and y, x origin 0. Outputs whose block has no nonzero voxel are zero vectors. | |
| ## How it was made | |
| - **Lasagna field**: [villa](https://github.com/ScrollPrize/villa) at commit `db3dbea26185a9e09ab997bfa21a2a00477f0d56`, | |
| `lasagna-preprocess predict3d`, tile size 192, with checkpoint `model_current.pt` of | |
| [`scrollprize/lasagna`](https://huggingface.co/scrollprize/lasagna) at revision `de45f01147adea9f4ad0f621b1ccd885a1f03760` | |
| (sha256 `4cc0c40a846b296769b9f7982470559078662a1a0b0e1f74ff41ed9341c0d600`); the run's `inference.json` records the | |
| villa checkout as having local changes (`repository.dirty`). | |
| - **Spiral fits**: villa's spiral fit at commit `e2254ebe0168a232e5ba7b94997c749b4e6a4544`, eager mode, 30,000 steps, | |
| seeds 1 to 3, on villa lasagna fields of the same scan, around a center line made with the | |
| [Center Finder](https://huggingface.co/datasets/SharpMeow/vesuvius-center-lines) (PHerc0172) or the open-data | |
| umbilicus (PHerc0332). | |
| - **m7**: [`scrollprize/surface_m7_nnunet`](https://huggingface.co/scrollprize/surface_m7_nnunet) at revision | |
| `cd83ccf7e1d58b70bd9074a9a7e158cd2c894f20`, run with villa at commit `e0bbb8b40a2db58b1d71864f286eb85717e59e64` | |
| (the plans' normalization, a 192-voxel halo, threshold 0.2). | |
| - **dinovol**: [`scrollprize/dinovol_8um_ps192_102126`](https://huggingface.co/scrollprize/dinovol_8um_ps192_102126), | |
| checkpoint `teacher_step_640000.pt` (sha256 `2354c493f050070e17ea0cda5cd3c1a6ebdde19a70a9c28bc9d368b80faabb86`), | |
| villa's `dinovol` code at commit `4ea26d9cbfc1778c68e0875c21c1f670922deba1`, robust normalization with fixed | |
| volume-wide statistics, a 192-voxel tile lattice at stride 96. | |
| - All runs used Hugging Face Jobs. The files were copied server side from the runs' working storage, byte for byte. | |
| ## License and credit | |
| The scans are Vesuvius Challenge open data, licensed [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/); | |
| everything here is derived from them and carries the same terms. Please credit the Vesuvius Challenge and its | |
| scanning partners for the scans, the villa authors for the lasagna, spiral fit, m7 and dinovol models and code, and | |
| this bucket. Made by Chase Hendrick, Independent Researcher. | |
Xet Storage Details
- Size:
- 9.86 kB
- Xet hash:
- 2a64fb06d2e3f8f7ee45daf5688cfa5b9c2305f7b0375ee611a513a309dc011e
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.