{"slug": "fragment-aware-vision-transformers-for-fresco-fragment-style-classification", "title": "Fragment-Aware Vision Transformers for Fresco-Fragment Style Classification", "summary": "A progressive transformer-based framework for fresco-fragment style classification improved accuracy from 0.604 to 0.656 and macro-F1 from 0.596 to 0.648 on the CLEOPATRA dataset over a standard ViT-B/16 baseline, according to an arXiv paper (2609.21012v1). The method combines foreground-guided masking, inpainting-based geometric regularisation, and a supervised contrastive objective using Kullback-Leibler similarity, ensembled via a learnable logit combination. On POMPAAF the ensemble outperformed the best single branch in four of six fragmentation settings, while a more complex graph-fusion variant matched the simple ensemble on POMPAAF and gave only a small, dataset-specific gain on CLEOPATRA.", "body_md": "arXiv:2609.21012v1 Announce Type: new \nAbstract: Artistic style classification is usually studied on complete artworks, where models can exploit global composition, spatial organisation, and iconographic structure. In archaeological settings, however, artworks often survive only as fragmented remains, forcing recognition from incomplete, irregular, and context-limited visual evidence. We study fresco-fragment style classification using a progressive transformer-based framework. Starting from a ViT-B/16 baseline, we introduce foreground-guided masking to suppress background-only tokens, inpainting-based geometric regularisation to align irregular fragment supports with the ViT patch grid, and a supervised contrastive objective that operates on predictive distributions through a Kullback-Leibler similarity and consistently improves every branch. We combine the branches with a deliberately simple learnable logit ensemble. Experiments on CLEOPATRA and POMPAAF show that fragment-aware modelling improves over the standard ViT baseline, with the ensemble increasing accuracy from 0.604 to 0.656 and macro-F1 from 0.596 to 0.648 on CLEOPATRA, and outperforming the best single branch in four of six fragmentation settings on POMPAAF. We additionally evaluate a more complex graph-fusion variant and find that it matches the simple ensemble on POMPAAF while offering only a small, dataset-specific gain on CLEOPATRA, which does not justify its added complexity. Beyond these empirical gains, our contribution is twofold: a distribution-level contrastive objective that consistently sharpens single-branch recognition, and an interpretability analysis that verifies the models exploit genuine painted evidence, while quantifying that the inpainting-based branch draws part of its attribution from the synthesised surround.", "url": "https://wpnews.pro/news/fragment-aware-vision-transformers-for-fresco-fragment-style-classification", "canonical_source": "https://arxiv.org/abs/2609.21012", "published_at": "2026-09-21 04:00:00+00:00", "updated_at": "2026-09-21 04:26:50.984629+00:00", "lang": "en", "topics": ["computer-vision", "machine-learning", "ai-research", "neural-networks"], "entities": ["CLEOPATRA", "POMPAAF", "ViT-B/16", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/fragment-aware-vision-transformers-for-fresco-fragment-style-classification", "markdown": "https://wpnews.pro/news/fragment-aware-vision-transformers-for-fresco-fragment-style-classification.md", "text": "https://wpnews.pro/news/fragment-aware-vision-transformers-for-fresco-fragment-style-classification.txt", "jsonld": "https://wpnews.pro/news/fragment-aware-vision-transformers-for-fresco-fragment-style-classification.jsonld"}}