Virtually Unrolling the Herculaneum Papyri by Diffeomorphic Spiral Fitting

TL;DR

The paper tackles the challenge of reading heavily damaged Herculaneum Papyri by proposing a top-down, model-based virtual unrolling method. It fits a diffeomorphic transform to map a canonical Archimedean spiral surface to noisy CT-based predictions, ensuring a single continuous 2D sheet even through missing regions. Key innovations include converting noisy predictions into sparse, trackable paths, a multi-stage, invertible transform (per-slice affine, integrated flow, gap scaling), and robust losses that align normals, winding spacing, and path distances. Evaluations on PHerc. Paris 4 and PHerc. 172 show improved geometric accuracy and usable unrolled surfaces with visible ink, outperforming prior automated approaches and enabling potential text extraction from previously unread scrolls.

Abstract

The Herculaneum Papyri are a collection of rolled papyrus documents that were charred and buried by the famous eruption of Mount Vesuvius. They promise to contain a wealth of previously unseen Greek and Latin texts, but are extremely fragile and thus most cannot be unrolled physically. A solution to access these texts is virtual unrolling, where the papyrus surface is digitally traced out in a CT scan of the scroll, to create a flattened representation. This tracing is very laborious to do manually in gigavoxel-sized scans, so automated approaches are desirable. We present the first top-down method that automatically fits a surface model to a CT scan of a severely damaged scroll. We take a novel approach that globally fits an explicit parametric model of the deformed scroll to existing neural network predictions of where the rolled papyrus likely passes. Our method guarantees the resulting surface is a single continuous 2D sheet, even passing through regions where the surface is not detectable in the CT scan. We conduct comprehensive experiments on high-resolution CT scans of two scrolls, showing that our approach successfully unrolls large regions, and exceeds the performance of the only existing automated unrolling method suitable for this data.

Figures (17)

• Figure 1: Our approach ingests noisy volumetric U-Net predictions of papyrus surface in a CT scan of a rolled scroll, and virtually unrolls it into a continuous 2D sheet surface. It fits a parametric model of an idealized scroll and a diffeomorphic transform given as the integral of a velocity field.
• Figure 2: (a--c) Cross-section, longwise section and volume rendering of PHerc. Paris 4, from the micro-CT scan in EduceLab-Scrolls parsons23dataset. The original scroll is highly compressed and distorted, and it is difficult to discern where the original 2D papyrus surface passes. (d) State-of-the-art segmentation of surface voxels (red) overlaid on the scroll. Note there are breaks in certain regions, as well as falsely joined sheets and spurious fragments. The predictions do not allow directly extracting a manifold surface. (e) Representative examples of datasets used in prior works on virtual unrolling/unfolding (CT-OCR-2022 polevoy22ctocr used in kulagin24icdar; Papyrus L/El227b/3-pU from klenert25wacvw; Papryus L/El227b/1-pC from mahnke20jch); compare with the complexity of PHerc. Paris 4 seen in (a--d).
• Figure 3: Cross-sections of PHerc. Paris 4 with the spiral overlaid in red, before optimization when the diffeomorphism is an identity transform (top-left inset); and after optimization. During fitting the spiral adapts to follow the distorted scroll surface. Far right: Undeformed cross-section, given by transforming the scan with the inverse diffeomorphism. The distorted windings become near-circular.
• Figure 4: Top: Visualization of the papyrus surface extracted from PHerc. Paris 4 by our method. The right-hand end corresponds to the center of the rolled scroll; the wavy edge at the top reflects the increasing radius of windings further from the center. Bottom: Ink predictions using the TimeSformer bertasius21icml model from nader23gpw, applied to the virtually unrolled volume. Note the rows of Greek characters (black) visible throughout the extracted surface. Best viewed with zoom.
• Figure 5: Top: deformed spiral overlaid on PHerc. 172. Bottom: Close-ups of the unrolled surface. Ink is visible as light areas; several Greek letters can be discerned. The full output of our method is in the appendix.