Unified and Multimodal Segmentation: Foundations, Methods, and Applications

Dear Colleagues,

Segmentation plays a pivotal role in computer vision, serving as a bridge between low-level perception and high-level visual understanding. With the rapid advancement of artificial intelligence and multimodal sensing technologies, segmentation has evolved from task-specific pipelines to unified frameworks capable of handling diverse modalities such as RGB, depth, thermal, CT, and MRI. These developments have opened new possibilities for applications across healthcare, industry, and general visual perception.

However, despite recent progress, current segmentation systems still face significant challenges in generalization, robustness, and interpretability. Traditional models often rely heavily on modality-specific supervision, lack adaptability to missing or corrupted modalities, and struggle to scale across heterogeneous domains. The emergence of foundation and prompt-driven models, self-supervised learning, and generative augmentation now provides a transformative opportunity to build truly unified segmentation systems that can learn and adapt seamlessly across multiple visual tasks and modalities.

This Special Issue aims to gather a diverse and complementary set of contributions that advance the theory and practice of unified and multimodal segmentation. We welcome papers that explore new frameworks, learning paradigms, and applications in areas including but not limited to multimodal data fusion, contrastive representation learning, self-supervised segmentation, cross-domain adaptation, and explainable visual reasoning.

By integrating advances from AI4Health and AI4Industry, this Special Issue seeks to promote dialogue between fundamental research and real-world deployment, fostering progress toward general, scalable, and intelligent segmentation systems that impact both science and society.

Dr. Xiaoqi Zhao
Dr. Youwei Pang
Dr. Kailai Zhou
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. Journal of Imaging 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 1800 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
• multimodal learning
• unified vision models
• generative AI
• AI4Health
• AI4Industry
• self-supervised learning
• visual understanding
• foundation models
• context-aware perception