Wavelet-based fourier information interaction with frequency diffusion adjustment for underwater image restoration

C Zhao, W Cai, C Dong, C Hu - Proceedings of the IEEE …, 2024‏ - openaccess.thecvf.com
Underwater images are subject to intricate and diverse degradation inevitably affecting the
effectiveness of underwater visual tasks. However most approaches primarily operate in the …

Docres: a generalist model toward unifying document image restoration tasks

J Zhang, D Peng, C Liu, P Zhang… - Proceedings of the …, 2024‏ - openaccess.thecvf.com
Document image restoration is a crucial aspect of Document AI systems as the quality of
document images significantly influences the overall performance. Prevailing methods …

Visual Text Meets Low-level Vision: A Comprehensive Survey on Visual Text Processing

Y Shu, W Zeng, Z Li, F Zhao, Y Zhou - arxiv preprint arxiv:2402.03082, 2024‏ - arxiv.org
Visual text, a pivotal element in both document and scene images, speaks volumes and
attracts significant attention in the computer vision domain. Beyond visual text detection and …

High-fidelity document stain removal via a large-scale real-world dataset and a memory-augmented transformer

M Li, H Sun, Y Lei, X Zhang, Y Dong, Y Zhou… - arxiv preprint arxiv …, 2024‏ - arxiv.org
Document images are often degraded by various stains, significantly impacting their
readability and hindering downstream applications such as document digitization and …

[PDF][PDF] A comprehensive survey on diffusion models and their applications

MM Ahsan, S Raman, Y Liu, Z Siddique - Preprints, August, 2024‏ - preprints.org
Diffusion Models (DMs) are probabilistic models that create realistic samples by simulating
the diffusion process, gradually adding and removing noise from data. These models have …

Generate, transform, and clean: the role of GANs and transformers in palm leaf manuscript generation and enhancement

N Thuon, J Du, Z Zhang, J Ma, P Hu - International Journal on Document …, 2024‏ - Springer
Palm leaf manuscripts offer a rich source of data critical for document analysis tasks,
including character, word, and text analysis. However, their cleaning and denoising present …

Reproducing the Past: A Dataset for Benchmarking Inscription Restoration

S Zhu, H Xue, N Nie, C Zhu, H Liu, P Fang - Proceedings of the 32nd …, 2024‏ - dl.acm.org
Inscriptions on ancient steles, as carriers of culture, encapsulate the humanistic thoughts
and aesthetic values of our ancestors. However, these relics often deteriorate due to …

Textdiff: Mask-guided residual diffusion models for scene text image super-resolution

B Liu, Z Yang, P Wang, J Zhou, Z Liu, Z Song… - arxiv preprint arxiv …, 2023‏ - arxiv.org
The goal of scene text image super-resolution is to reconstruct high-resolution text-line
images from unrecognizable low-resolution inputs. The existing methods relying on the …

Seal2real: prompt prior learning on diffusion model for unsupervised document seal data generation and realisation

J Huang, Y Liu, Y Huang, S Chen - arxiv preprint arxiv:2310.00546, 2023‏ - arxiv.org
In document processing, seal-related tasks have very large commercial applications, such
as seal segmentation, seal authenticity discrimination, seal removal, and text recognition …

Naf-dpm: A nonlinear activation-free diffusion probabilistic model for document enhancement

G Cicchetti, D Comminiello - arxiv preprint arxiv:2404.05669, 2024‏ - arxiv.org
Real-world documents may suffer various forms of degradation, often resulting in lower
accuracy in optical character recognition (OCR) systems. Therefore, a crucial preprocessing …