Neural Networks and Deep Learning in Computer Vision

Dear Colleagues,

There is no question that the driving force in computer vision in recent years has been the rapid development of deep learning, and neural networks in particular. Applications like autonomous driving, disease diagnosis, text-to-image generators, and many more were unimaginable just ten years ago. The ability of neural networks to model and discover complex dependencies hidden in data, and to mimic the way the human brain works, paves the way for these, and many more, applications into our daily lives. Many practical open questions in computer vision, on the other hand, stimulate researchers to seek new and better approaches in deep learning and neural networks. Despite the optimistic view, many challenges remain. The recent development of the Forward-Forward algorithm is one such instance, which raises the question of whether backpropagation will still be useful in the future. Typical problems in computer vision include determining if a specific neural network architecture or training approach is appropriate for a given application, such as the optimal deep learning approach for neural radiation fields. The metaverse creates a whole new set of computer-vision-related issues, including the need for specialized neural networks that can execute in real time and provide a realistic experience to the user.

This Special Issue aims to provide a platform for original contributions related to emerging deep learning architectures for computer vision as well as computer vision applications, opening the door for new neural network architectures. Research topics in this Special Issue include, but are not limited to:

• Emerging neural network architectures with applications to computer vision, e.g., FocalNet, declarative neural networks, diffusion models, graph-convolutional neural networks;
• Bayesian deep learning for computer vision;
• Self-supervised/contrastive/generative learning;
• Semantically guided analysis for image and video, e.g., semantically guided compression/text-to-image/analysis;
• Neural radiance fields and 3D reconstruction from single or multiple views;
• Neural models for compressible, animatable, relightable 3D human face and body models;
• Three-dimensional image and video analysis: segmentation, recognition, motion analysis and prediction.

Dr. Krasimir Tonchev
Dr. Agata Manolova
Prof. Dr. Petia Georgieva
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 250 words) can be sent to the Editorial Office for assessment.

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.