Computer Vision Based on Generative AI Models

Dear Colleagues,

Recent advances in generative AI models, particularly diffusion models, generative adversarial networks (GANs), and large vision-language models (VLMs), have significantly reshaped the landscape of computer vision. Unlike traditional discriminative approaches, generative models enable data synthesis, image-to-image translation, 3D reconstruction, scene simulation, and cross-modal understanding. This opens transformative opportunities in applications ranging from medical imaging and remote sensing to autonomous driving, surveillance, and digital twin systems.

Despite these advances, many challenges remain. Key open problems include improving model efficiency and interpretability, addressing domain adaptation and bias, enhancing robustness under real-world constraints (e.g., low-light conditions, occlusions, or adverse weather), and ensuring trustworthy AI governance. This Special Issue aims to bring together cutting-edge contributions that explore how generative AI models can push forward the boundaries of computer vision in theory, algorithms, and applications.

This Special Issue seeks original research papers, surveys, and visionary perspectives on novel methods, systems, and applications of generative AI in computer vision. Topics include (but are not limited to):

• Generative Models for Vision;
• Diffusion models, GANs, VAEs, autoregressive and transformer-based generative models;
• Vision-Language Models (e.g., CLIP, BLIP, LLaVA) for generative perception and reasoning;
• Applications;
• Image/video restoration, enhancement, and dehazing;
• 3D/4D scene reconstruction and view synthesis;
• Human pose, gesture, and facial expression generation/recognition;
• Generative data augmentation for imbalanced or low-resource vision tasks;
• Cross-modal translation (e.g., text-to-image, image-to-text, sketch-to-image);
• Synthetic dataset generation for autonomous driving, medical imaging, and industrial inspection;
• Trustworthy and Efficient Generative Vision Systems;
• Lightweight generative architectures for edge AI;
• Robustness, explainability, and fairness of generative vision models;
• Evaluation metrics for generative computer vision;
• Federated and privacy-preserving generative learning;
• Emerging Trends;
• Generative digital twins and metaverse applications;
• Generative foundation models for large-scale vision problems;
• Integration of generative AI with reinforcement learning and robotics.

Dr. Yi-Zeng Hsieh
Dr. De-Yuan Huang
Dr. Chen-Chiung Hsieh
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. 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.

• generative models for vision
• diffusion models
• vision-language models