Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Preserving Old Memories in Vivid Detail: Human-Interactive Photo Restoration Framework

Seung-Yeon Back, Geonho Son, Dahye Jeong, Eunil Park, Simon S. Woo

TL;DR

This work presents the AI-based photo restoration framework composed of multiple stages, where each stage is tailored to enhance and restore specific types of photo damage, accelerating and automating the photo restoration process.

Abstract

Photo restoration technology enables preserving visual memories in photographs. However, physical prints are vulnerable to various forms of deterioration, ranging from physical damage to loss of image quality, etc. While restoration by human experts can improve the quality of outcomes, it often comes at a high price in terms of cost and time for restoration. In this work, we present the AI-based photo restoration framework composed of multiple stages, where each stage is tailored to enhance and restore specific types of photo damage, accelerating and automating the photo restoration process. By integrating these techniques into a unified architecture, our framework aims to offer a one-stop solution for restoring old and deteriorated photographs. Furthermore, we present a novel old photo restoration dataset because we lack a publicly available dataset for our evaluation.

Preserving Old Memories in Vivid Detail: Human-Interactive Photo Restoration Framework

TL;DR

This work presents the AI-based photo restoration framework composed of multiple stages, where each stage is tailored to enhance and restore specific types of photo damage, accelerating and automating the photo restoration process.

Abstract

Photo restoration technology enables preserving visual memories in photographs. However, physical prints are vulnerable to various forms of deterioration, ranging from physical damage to loss of image quality, etc. While restoration by human experts can improve the quality of outcomes, it often comes at a high price in terms of cost and time for restoration. In this work, we present the AI-based photo restoration framework composed of multiple stages, where each stage is tailored to enhance and restore specific types of photo damage, accelerating and automating the photo restoration process. By integrating these techniques into a unified architecture, our framework aims to offer a one-stop solution for restoring old and deteriorated photographs. Furthermore, we present a novel old photo restoration dataset because we lack a publicly available dataset for our evaluation.

Paper Structure

This paper contains 7 sections, 5 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Architecture
  4. Dataset
  5. Experiments
  6. Tool Usage
  7. Conclusion

Figures (5)

  • Figure 1: Overview of our Restoration Framework showing the process of restoring a given damaged image. The blue lines represent the workflow of the framework, while the orange lines indicate user feedback. The image with red dotted lines connecting different components illustrates the inputs and outputs for each module of the framework.
  • Figure 2: Comparison with baseline restoration frameworks. Ours-A represents the results when our pipeline is utilized automatically with the given mask, while Ours-HI denotes the results when our pipeline is manually used with human interaction.
  • Figure 3: Additional results of restoring real-world old photos using our framework with human interaction.
  • Figure 4: Comparison of human evaluation ratios across two questions.
  • Figure 5: Restoration interface implementing our proposed framework.