Advances in Computer Vision for Image Segmentation and Scene Understanding

Dear Colleagues,

Recent breakthroughs in artificial intelligence, large-scale datasets, and computational imaging have significantly advanced research in image segmentation and scene understanding. These tasks serve as core components in numerous applications, including autonomous driving, medical image analysis, robotics, remote sensing, and intelligent surveillance. However, despite remarkable progress, the field still faces major challenges related to robustness, cross-domain generalization, multi-sensor integration, long-range contextual modeling, and the reliable deployment of perception systems in dynamic real-world environments. In particular, handling degraded or low-quality visual data—such as low-light images, motion blur, occlusions, noise, and adverse weather—remains a pressing problem that demands new algorithmic innovations.

This Special Issue aims to highlight the latest advances in image segmentation and scene understanding, with a special focus on modern AI-driven techniques. We welcome research on multimodal alignment and fusion strategies that integrate RGB images, depth maps, LiDAR point clouds, infrared imagery, radar signals, and textual or audio modalities, enabling richer and more consistent semantic interpretation. Contributions on multi-scale representation learning, hierarchical scene modeling, structural reasoning, and fine-grained semantic understanding are also strongly encouraged.

Emerging paradigms such as transformer-based architectures, diffusion models, large vision–language foundation models, and unified multimodal frameworks have opened new opportunities for holistic scene perception. Studies exploring innovative model architectures, efficient training strategies, domain adaptation and generalization methods, and advanced inference pipelines fall within the scope of this Special Issue. Furthermore, as visual systems are increasingly deployed in safety-critical and resource-constrained environments, we particularly encourage work addressing interpretability, robustness, fairness, uncertainty estimation, and trustworthy AI for segmentation and scene understanding.

While the existing literature has achieved substantial progress in segmentation algorithms and scene-level reasoning frameworks, a gap remains between theoretical developments and practical deployment across diverse and challenging environments. By bringing together contributions from computer vision, machine learning, multimodal sensing, and intelligent systems, this Special Issue seeks to provide a comprehensive platform that showcases cutting-edge advances and identifies emerging challenges and opportunities for the next generation of perception technologies.

We look forward to receiving your valuable contributions.

Prof. Dr. Zhikui Chen
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 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.

• image segmentation
• scene understanding
• multi-scale representation
• multimodal fusion
• diffusion models
• multimodal learning
• multimodal alignment
• robust and trustworthy AI
• representation learning