# Zero-Shot SAM2 Segmentation and Vision Transformer-Based Recognition of Elamite Cuneiform Symbols from Degraded Tablet Images

arXiv: https://arxiv.org/abs/2608.18544 · HTML: https://mlsanity.com/paper/2608.18544/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 55 references that are in the map.
- First posted on arXiv 19 Aug 2026.

## Builds on (most cited first, 20 of 20)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors](https://mlsanity.com/paper/2207.02696/) (2022)
- [Unsupervised Structure-Texture Separation Network for Oracle Character Recognition](https://mlsanity.com/paper/2205.06549/) (2022)
- [Recognizing Cuneiform Signs Using Graph Based Methods](https://mlsanity.com/paper/1802.05908/) (2018)
- [Recognition of Oracle Bone Inscriptions by using Two Deep Learning Models](https://mlsanity.com/paper/2105.00777/) (2021)
- [Deep learning classification of large-scale point clouds: A case study on cuneiform tablets](https://mlsanity.com/paper/2202.10851/) (2022)
- [MapSAM: Adapting Segment Anything Model for Automated Feature Detection in Historical Maps](https://mlsanity.com/paper/2411.06971/) (2024)
- [CNN based Cuneiform Sign Detection Learned from Annotated 3D Renderings and Mapped Photographs with Illumination Augmentation](https://mlsanity.com/paper/2308.11277/) (2023)
- [Ancient Script Image Recognition and Processing: A Review](https://mlsanity.com/paper/2506.19208/) (2025)
- [ProtoSnap: Prototype Alignment for Cuneiform Signs](https://mlsanity.com/paper/2502.00129/) (2025)
- [DeepScribe: Localization and Classification of Elamite Cuneiform Signs Via Deep Learning](https://mlsanity.com/paper/2306.01268/) (2023)
- [Deep Aramaic: Towards a Synthetic Data Paradigm Enabling Machine Learning in Epigraphy](https://mlsanity.com/paper/2310.07310/) (2023)
- [A Recursive Encoding for Cuneiform Signs](https://mlsanity.com/paper/2406.17283/) (2024)
- [Advanced Deep Learning Approaches for Automated Recognition of Cuneiform Symbols](https://mlsanity.com/paper/2505.04678/) (2025)

## Abstract (arXiv)

Automated recognition of ancient cuneiform script poses a compound signal-degradation problem: the three-dimensional relief of clay tablets creates spatially varying illumination and cast shadows, surface erosion introduces structured noise that overlaps with genuine sign impressions, and severe class imbalance across 141 sign categories undermines classifier reliability. We introduce EpigraphNet, a segmentation-guided transformer pipeline evaluated on the Persepolis Fortification Archive. From 1,239 annotated tablet images, brightness-adaptive morphological preprocessing and zero-shot SAM2-Large segmentation generate clean binary symbol masks, which a fine-tuned Vision Transformer (ViT-B/16) with inverse-frequency class weighting then classifies. EpigraphNet reaches 86.41% top-1 accuracy on a 132-class benchmark, a 17.21 percentage-point gain over the strongest CNN baseline (ResNet-101, 69.20%) and 5.31-12.91% over four modern backbones (DeiT-B/16, Swin-B, ConvNeXt-B, EfficientNet-B4) under identical conditions. The full pipeline runs at approximately 18 ms per sign on an NVIDIA A100 GPU. A lower Spearman correlation between sign frequency and per-class performance indicates more balanced recognition across frequent and rare classes. Implementation is available at: github.com/r11up/sam-guided-vit