Progressive enhancement and restoration for mural images under low-light and defected conditions based on multi-receptive field strategy

Read original: arXiv:2405.08245 - Published 7/18/2024 by Xiameng Wei, Binbin Fan, Ying Wang, Yanxiang Feng, Laiyi Fu

🌐

Overview

Ancient murals are valuable cultural heritage that provide insights into ancient civilizations
These murals have suffered continuous damage over time due to environmental factors and inadequate protection
Capturing high-quality images of damaged murals is challenging, especially in low-light conditions, which hinders further restoration efforts
To address these challenges, the researchers propose a two-stage restoration model called MER (Mural Enhancement and Restoration net) to enhance the visual quality of damaged murals captured in low-light conditions and facilitate their restoration

Plain English Explanation

Ancient murals, or wall paintings, are incredibly valuable for understanding the beliefs, customs, and history of ancient civilizations. These murals often depict religious ceremonies, folklore, and other cultural practices, making them a crucial part of our archaeological heritage. However, over time, these murals have suffered significant damage due to oxidation, mold, and other environmental factors, as well as a lack of adequate protection.

One of the key challenges in restoring these murals is the difficulty in capturing high-quality images, especially in low-light conditions, which are common for indoor murals. The poor visibility of the damaged areas can make it challenging for experts to identify the extent of the damage and plan effective restoration strategies.

To address this problem, the researchers have developed a two-stage restoration model called MER (Mural Enhancement and Restoration net). This model not only enhances the visual quality of the restored images but also achieves impressive results when evaluated using various metrics, outperforming other competing methods.

Additionally, the researchers have launched a website dedicated to the restoration of ancient mural paintings, utilizing the proposed MER model to assist in the conservation and preservation of this invaluable cultural heritage.

Technical Explanation

The researchers have developed a two-stage MER (Mural Enhancement and Restoration net) model to address the challenges of restoring ancient murals captured in low-light conditions. The first stage of the model focuses on enhancing the visual quality of the images, improving factors such as contrast, brightness, and detail preservation.

This enhancement stage leverages techniques like multi-feature reconstruction and exposure bracketing to create a high-quality, visually appealing image from the low-light original. The second stage of the model then tackles the actual restoration of the damaged areas, leveraging techniques like inhomogeneous illumination enhancement and detail-preserving restoration to identify and repair the damaged regions.

The researchers have evaluated the performance of the MER model using various metrics and have found it to outperform other competing methods in terms of both visual quality and restoration accuracy. This suggests that the proposed two-stage approach is an effective solution for addressing the challenges of restoring ancient murals captured in low-light conditions.

Critical Analysis

The researchers have presented a compelling solution for the restoration of ancient murals, addressing the key challenges of low-light capture and continuous damage to these valuable cultural artifacts. However, it is important to note that the effectiveness of the MER model may be limited to certain types of damage or lighting conditions, and its performance may vary depending on the specific characteristics of the murals being restored.

Additionally, the researchers have not explored the potential long-term implications of their proposed model, such as the sustainability of the restoration process or the impact on the overall preservation of ancient murals. Further research may be needed to address these broader concerns and ensure that the MER model is truly a comprehensive solution for the conservation of this cultural heritage.

It is also worth considering the potential ethical implications of using AI-powered restoration techniques, as there may be concerns about the authenticity and integrity of the restored images. The researchers should continue to engage with relevant stakeholders, including cultural heritage experts and community members, to ensure that the MER model is developed and deployed in a responsible and transparent manner.

Conclusion

The researchers have developed a promising two-stage MER (Mural Enhancement and Restoration net) model to address the challenges of restoring ancient murals captured in low-light conditions. By enhancing the visual quality of the images and then tackling the actual restoration of damaged areas, the MER model has demonstrated impressive results in relevant metric evaluations, outperforming other competing methods.

The launch of a dedicated website for the restoration of ancient mural paintings, utilizing the proposed MER model, is a commendable step towards preserving this invaluable cultural heritage. However, further research is needed to address the potential limitations and broader implications of this approach, ensuring that the restoration process is sustainable and respects the integrity of the original artifacts.

Overall, the MER model represents a significant step forward in the field of ancient mural restoration, and its continued development and refinement could have far-reaching impacts on the preservation of our shared cultural legacy.

This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!

Follow @aimodelsfyi on 𝕏 →

Related Papers

🌐

Progressive enhancement and restoration for mural images under low-light and defected conditions based on multi-receptive field strategy

Xiameng Wei, Binbin Fan, Ying Wang, Yanxiang Feng, Laiyi Fu

Ancient murals are valuable cultural heritage with great archaeological value. They provide insights into ancient religions, ceremonies, folklore, among other things through their content. However, due to long-term oxidation and inadequate protection, ancient murals have suffered continuous damage, including peeling and mold etc. Additionally, since ancient murals were typically painted indoors, the light intensity in images captured by digital devices is often low. The poor visibility hampers the further restoration of damaged areas. To address the escalating damage to ancient murals and facilitate batch restoration at archaeological sites, we propose a two-stage restoration model with automatic defect area detection strategy which called MER(Mural Enhancement and Restoration net) for ancient murals that are damaged and have been captured in low light. Our two-stage model not only enhances the visual quality of restored images but also achieves commendable results in relevant metric evaluations compared with other competitors. Furthermore, we have launched a website dedicated to the restoration of ancient mural paintings, utilizing the proposed model. Code is available at https://gitee.com/bbfan2024/MER.git.

7/18/2024

🛠️

Line Drawing Guided Progressive Inpainting of Mural Damage

Luxi Li, Qin Zou, Fan Zhang, Hongkai Yu, Long Chen, Chengfang Song, Xianfeng Huang, Xiaoguang Wang, Qingquan Li

Mural image inpainting is far less explored compared to its natural image counterpart and remains largely unsolved. Most existing image-inpainting methods tend to take the target image as the only input and directly repair the damage to generate a visually plausible result. These methods obtain high performance in restoration or completion of some pre-defined objects, e.g., human face, fabric texture, and printed texts, etc., however, are not suitable for repairing murals with varying subjects and large damaged areas. Moreover, due to discrete colors in paints, mural inpainting may suffer from apparent color bias. To this end, in this paper, we propose a line drawing guided progressive mural inpainting method. It divides the inpainting process into two steps: structure reconstruction and color correction, implemented by a structure reconstruction network (SRN) and a color correction network (CCN), respectively. In structure reconstruction, SRN utilizes the line drawing as an assistant to achieve large-scale content authenticity and structural stability. In color correction, CCN operates a local color adjustment for missing pixels which reduces the negative effects of color bias and edge jumping. The proposed approach is evaluated against the current state-of-the-art image inpainting methods. Qualitative and quantitative results demonstrate the superiority of the proposed method in mural image inpainting. The codes and data are available at https://github.com/qinnzou/mural-image-inpainting.

7/8/2024

OneRestore: A Universal Restoration Framework for Composite Degradation

Yu Guo, Yuan Gao, Yuxu Lu, Huilin Zhu, Ryan Wen Liu, Shengfeng He

In real-world scenarios, image impairments often manifest as composite degradations, presenting a complex interplay of elements such as low light, haze, rain, and snow. Despite this reality, existing restoration methods typically target isolated degradation types, thereby falling short in environments where multiple degrading factors coexist. To bridge this gap, our study proposes a versatile imaging model that consolidates four physical corruption paradigms to accurately represent complex, composite degradation scenarios. In this context, we propose OneRestore, a novel transformer-based framework designed for adaptive, controllable scene restoration. The proposed framework leverages a unique cross-attention mechanism, merging degraded scene descriptors with image features, allowing for nuanced restoration. Our model allows versatile input scene descriptors, ranging from manual text embeddings to automatic extractions based on visual attributes. Our methodology is further enhanced through a composite degradation restoration loss, using extra degraded images as negative samples to fortify model constraints. Comparative results on synthetic and real-world datasets demonstrate OneRestore as a superior solution, significantly advancing the state-of-the-art in addressing complex, composite degradations.

7/11/2024

Training-Free Large Model Priors for Multiple-in-One Image Restoration

Xuanhua He, Lang Li, Yingying Wang, Hui Zheng, Ke Cao, Keyu Yan, Rui Li, Chengjun Xie, Jie Zhang, Man Zhou

Image restoration aims to reconstruct the latent clear images from their degraded versions. Despite the notable achievement, existing methods predominantly focus on handling specific degradation types and thus require specialized models, impeding real-world applications in dynamic degradation scenarios. To address this issue, we propose Large Model Driven Image Restoration framework (LMDIR), a novel multiple-in-one image restoration paradigm that leverages the generic priors from large multi-modal language models (MMLMs) and the pretrained diffusion models. In detail, LMDIR integrates three key prior knowledges: 1) global degradation knowledge from MMLMs, 2) scene-aware contextual descriptions generated by MMLMs, and 3) fine-grained high-quality reference images synthesized by diffusion models guided by MMLM descriptions. Standing on above priors, our architecture comprises a query-based prompt encoder, degradation-aware transformer block injecting global degradation knowledge, content-aware transformer block incorporating scene description, and reference-based transformer block incorporating fine-grained image priors. This design facilitates single-stage training paradigm to address various degradations while supporting both automatic and user-guided restoration. Extensive experiments demonstrate that our designed method outperforms state-of-the-art competitors on multiple evaluation benchmarks.

7/19/2024