The

Vesuvius

Challenge JournalEdition of 2026-11-03

Last week

13 articles · 27 Oct 2026, 08:00 UTC — 03 Nov 2026, 08:00 UTC

Papyrology desk · 2026-11-03

A second PHerc. 306B area opens for collaborative reading

Following the PHerc. 306B prize announcement, a separate contributor shared a flattened mesh, traced lines, ink images and CT images for another area of that scroll. The post describes roughly nineteen square…

2 original sources · posting dates

Ink desk · 2026-11-03

Update: PHerc. 1447 candidate gains model-agreement and background checks

The signal-versus-noise discussion now includes quantitative checks on a C-shaped candidate in PHerc. 1447 segment 20250702235910. A contributor reports agreement between v8-in, trained without this scroll, and two ink_9um seeds, alongside seven…

5 original sources · posting dates

Segmentation desk · 2026-11-03

Winding-field work adds an interim OBJ-to-tifxyz conversion route

A request to import winding-field OBJ output into VC3D drew a concrete integration update: the developer said native tifxyz output was in progress and that a conversion script had been sent as…

2 original sources · posting dates

Ink desk · 2026-11-03

Raw-CT scan-volume model moves into a full Paris 4 training run

A contributor reported completing a revised three-dimensional scan-volume architecture that takes raw CT with global z-score normalization, rather than the altered inputs used by an earlier version. The post says tutorial-set training…

2 original sources · posting dates

Tools & data desk · 2026-11-02

An OpenGL canvas broadens vc3d’s visualization options

A developer reported extracting an OpenGL canvas from a fibers3d contribution into a more general vc3d canvas. The discussion identified surface normals and isometry as possible overlays, alongside device-accelerated visualization. A later…

5 original sources · posting dates

Ink desk · 2026-11-02

Faster unrolling still faces a readability bottleneck

A contributor described progress with unrolling and three-dimensional ink detection at 9.6 micrometres on Paris 4, while stressing that recent winding-model improvements had stalled. The same account estimated roughly two hours for…

5 original sources · posting dates

Papyrology desk · 2026-11-02

Promising strokes meet a familiar question: signal or noise?

A papyrology discussion returned to the difficulty of separating apparent letter shapes from segmentation artefacts and visual noise. After a participant asked whether a candidate was promising, another suggested checking the raw…

5 original sources · posting dates