Information Theory in Computer Vision and Artificial Intelligence
A special issue of Entropy (ISSN 1099-4300). This special issue belongs to the section "Information Theory, Probability and Statistics".
Deadline for manuscript submissions: closed (20 August 2023) | Viewed by 481
Special Issue Editors
Interests: computer vision; AI; pattern recognition; machine learning; image processing; visual analytics
Special Issue Information
Dear Colleagues,
Recent research on computer vision and artificial intelligence has driven the development of various new vision-based applications (medical image analysis, art conservation practices, autonomous driving, precision agriculture etc.). Despite remarkable performance reported in many of the latest neural network models, it becomes increasingly crucial to understand the uncertainty of these methods in order to fit them under the umbrella of a probabilistic framework. This could bring benefits to: (1) control the potential risks of using deep neural network (DNN) models by achieving insightful understanding of these methods; (2) design more reliable and explainable models by exploring their uncertainty; and (3) exploit heterogenous data sources and integrate multiple modalities with well-distributed probabilistic outputs.
Information theory has been applied in computer vision and artificial intelligence because it provides objective metrics to measure uncertainties. Many metrics, including mutual information, cross entropy, and KL divergence, have been widely used as loss terms for the convergence of typical DNN models. Recently, meta-learning methods (e.g., neural processes) are also proposed to connect the DNNs with probabilistic theory. This Special Issue is collecting cutting-edge research ideas on the use of information theory in computer vision and artificial intelligence. Topics of interest include, but are not limited to:
- Information-theoretic metrics in DNN loss designs, such as mutual information, KL divergence, and Wasserstein loss etc.
- Uncertainty analysis of DNNs.
- Probabilistic DNNs.
- Heterogenous-sources-based fusion in AI.
- Meta-learning methods.
- Explainable AI.
- Applications of CV and AI.
Dr. Hui Fang
Dr. Xiyao Liu
Guest Editors
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. Entropy is an international peer-reviewed open access monthly 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 2600 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
- computer vision
- image processing
- deep neural network
- uncertainty analysis
- meta-learning
- model fusion
- entropy
- information theory
