The IRMA Community
Newsletters
Research IRM
Click a keyword to search titles using our InfoSci-OnDemand powered search:
|
Advancements in Image Restoration Techniques: A Comprehensive Review and Analysis through GAN
|
|
Author(s): Rohit Rastogi (Dept. of CSE, ABES Engineering College, India), Vineet Rawat (Dept. of CSE, ABES Engineering College, India)and Sidhant Kaushal (Dept. of CSE, ABES Engineering College, India)
Copyright: 2025
Pages: 38
Source title:
Generative Artificial Intelligence and Ethics: Standards, Guidelines, and Best Practices
Source Author(s)/Editor(s): Loveleen Gaur (University of South Pacific, Fiji & Taylor's University, Malaysia)
DOI: 10.4018/979-8-3693-3691-5.ch003
Purchase
|
Abstract
Image restoration poses a formidable challenge in the field of computer vision, endeavoring to restore high-quality images from degraded or corrupted versions. This research paper conducts a comprehensive comparison of three prominent image restoration methodologies: GFP GAN, DeOldify, and MIRNet. GFP GAN, featuring a specialized GAN architecture designed for image restoration tasks, introduces an AI-centric approach. DeOldify, a deep learning-based method, focuses on colorizing and restoring old images using advanced AI techniques, while MIRNet offers a lightweight network specifically crafted for image restoration within an AI framework. The comparative analysis involves training and testing each method on a diverse dataset comprising both degraded and ground truth images. Employing a confusion matrix, precision, accuracy, recall, and other evaluation metrics are computed to comprehensively assess the performance of these AI-based methods. The matrix affords insights into the strengths and weaknesses of each AI-driven approach, providing a nuanced understanding of their respective performances. Experimental evaluations divulge the relative effectiveness of GFP GAN, DeOldify, and MIRNet in addressing image restoration challenges, encompassing issues such as noise, blur, compression artifacts, and other degradations, within the context of AI methodologies. The results not only illuminate the advantages and limitations of each AI-infused method but also serve as a valuable resource for researchers and practitioners in selecting the most suitable AI-driven approach for their unique image restoration requirements. In conclusion, this paper offers a thorough comparison of GFP GAN, DeOldify, and MIRNet in the domain of AI-driven image restoration, leveraging a confusion matrix to analyze precision, accuracy, and other pertinent parameters. Through a meticulous consideration of these AI-powered evaluation metrics, this study furnishes invaluable insights into the nuanced performance of these methodologies, facilitating informed decision-making in various AI-driven image restoration applications
Related Content
|
Pooja Dehankar, Susanta Das.
© 2025.
28 pages.
|
|
Anam Afaq, Meenu Chaudhary, Loveleen Gaur.
© 2025.
24 pages.
|
|
Rohit Rastogi, Vineet Rawat, Sidhant Kaushal.
© 2025.
38 pages.
|
|
Saranya M., Amutha B..
© 2025.
18 pages.
|
|
Saranya M., Amutha B..
© 2025.
18 pages.
|
|
Amrik Singh.
© 2025.
16 pages.
|
|
shikha Nagar, Anam Afaq, Shilpa Narula.
© 2025.
26 pages.
|
|
|