Image/Video Processing and Encoding for Contemporary Applications
A special issue of Electronics (ISSN 2079-9292). This special issue belongs to the section "Computer Science & Engineering".
Deadline for manuscript submissions: 31 May 2024 | Viewed by 4977
Special Issue Editor
Interests: deep learning; image/video signal processing
Special Issues, Collections and Topics in MDPI journals
Special Issue Information
Dear Colleagues,
Even though the topics of image/video signal processing and compression have been studied for many years, the research trend is evolving with recent emerging ideas and methods for various new applications and needs. For example, due to artificial intelligence (AI) advances, many improvements in image/video signal processing have been made to improve the quality and understanding of the images and video scenes. Also, there are increasing numbers of papers to apply machine learning and learning-based approaches to image and video encoding and image communication areas. These image/video processing and encoding with AI techniques show a state-of-the-art performance at diverse applications: autonomous driving, medical imaging, CCTV surveillance, factory inspection, image/video coding, and communication, etc.
The focus of this Special Issue is the state-of-the-art research related to the image/video processing and encoding of recent learning-based methods and/or novel approaches into diverse applications. Topics of interest include, but are not limited to:
- Image/video acquisition, representation, presentation, and display
- Image/video processing, filtering and transforms, analysis and synthesis
- Learning and understanding of image/video data
- Image/video compression, transmission, communication, and networking
- Image/video pre/post-processing, video restoration, and super-resolution, etc.
- Machine learning/deep learning schemes for image/video processing and coding
- Diverse image/video applications such as medical imaging, autonomous driving, etc.
Prof. Dr. Jitae Shin
Guest Editor
Manuscript Submission Information
Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.
Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.
Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.
Keywords
- image and video processing
- image and video coding
- machine learning/deep learning
- image/video applications
- image/video communication
Planned Papers
The below list represents only planned manuscripts. Some of these manuscripts have not been received by the Editorial Office yet. Papers submitted to MDPI journals are subject to peer-review.
Title: Detecting Forged Images and Deepfake Videos via Content Consistency Evaluation
Authors: Po-Chyi Su; Bo-Hong Huang; Tien-Ying Kuo
Affiliation: Dept. of Computer Science and Information Engineering, National Central University, Taoyuan, Taiwan
Abstract: Image inpainting and Deepfake can significantly change the meaning of imagery contents, so both are considered serious threats to the integrity of visual data. Observing that such manipulations may lead to content inconsistency in images and video frames, we propose a deep-learning-based forensic scheme in this research via evaluating content consistency for identifying forgery or affected areas in images and videos. Since the ways of tampering are diverse, it is impractical to collect enough tampered data for supervised learning. The proposed method avoids using tampered data of various kinds in the training process but employs the information of original/unaltered contents instead. A feature extraction neural network is trained for classifying imagery blocks or patches. The similarity measurement using the Siamese network to evaluate the consistency of patch pairs helps to locate tampered areas. For image manipulation detection, a segmentation network is employed to refine the manipulated regions further. In the cases of Deepfake video detection, facial regions are first located, and then the video's authenticity is determined by comparing the similarity of such regions between consecutive frames. Extensive tests are applied on publicly available datasets, encompassing images and videos with various manipulation operations. The experimental results demonstrate superior accuracy and stability compared to existing methods.