TL;DR
This paper introduces a comprehensive dataset and an innovative automated method for restoring severely damaged historical documents, significantly improving OCR accuracy and enabling better preservation of cultural heritage.
Contribution
It presents a new full-page HDR dataset and a three-stage restoration approach that combines automation with human collaboration, advancing the field of historical document restoration.
Findings
OCR accuracy improved from 46.83% to 84.05% with AutoHDR
Further enhancement to 94.25% through human-machine collaboration
AutoHDR outperforms existing methods in restoring severely damaged documents
Abstract
Historical documents represent an invaluable cultural heritage, yet have undergone significant degradation over time through tears, water erosion, and oxidation. Existing Historical Document Restoration (HDR) methods primarily focus on single modality or limited-size restoration, failing to meet practical needs. To fill this gap, we present a full-page HDR dataset (FPHDR) and a novel automated HDR solution (AutoHDR). Specifically, FPHDR comprises 1,633 real and 6,543 synthetic images with character-level and line-level locations, as well as character annotations in different damage grades. AutoHDR mimics historians' restoration workflows through a three-stage approach: OCR-assisted damage localization, vision-language context text prediction, and patch autoregressive appearance restoration. The modular architecture of AutoHDR enables seamless human-machine collaboration, allowing for…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
