Submission Deadline: 31 January 2027 View: 90 Submit to Special Issue
Assoc. Prof. Bo Yang
Email: boyang@uestc.edu.cn
Affiliation: School of Automation, University of Electronic Science and Technology of China, Chengdu, China
Research Interests: computer vision, surgical robotics, surgical (endoscopic) vision, medical image processing
Dr. Chao Liu
Email: liu@lirmm.fr
Affiliation: Department of Robotics, LIRMM, University of Montpellier—CNRS, Montpellier, France
Research Interests: visual augmentation and reconstruction, 3D reconstruction of deformable surface, haptics in human-machine interaction, multimodal sensor-based analysis of manipulation skills, surgical robot, medical image processing
Over the past decade or so, AI technologies based on deep learning have made remarkable progress, particularly in the fields of signal processing and computer vision. Methods based on deep learning are being developed and commercialized at an unprecedented rate, which is dramatically changing the way humans live, learn, and work. While data-driven deep learning continues to improve performance, this special issue also welcomes submissions exploring classical, knowledge-driven signal and vision processing methods. We are particularly interested in submissions that explore combining these two paradigms, as we believe that methods fusing data and knowledge could overcome the limitations imposed by the interplay between data, computation, and model architecture.
Computer vision has been one of the most dynamic areas of research in the field of deep learning since convolutional neural networks experienced a resurgence in popularity in 2010. Popular research topics include image and video synthesis and generation, 3D vision, visual language models, and multimodal learning. Computer vision is accelerating its transition from virtual perception to embodied interaction in the physical world, evolving from vision-language models (VLMs) to vision-language-action models (VLAs). In this context, AI must engage with more fundamental signals and hardware information. AI intersects with traditional control, automation, and robotics technologies in signal processing, resulting in the convergence of these disciplines.
In short, AI has progressed from recognizing the world through traditional vision tasks, such as classification and detection, to simulating the world through generative models, and finally, changing the world through embodied intelligence. The second edition of this special issue aims to document and advance these evolving trends, emphasizing breakthroughs in signal processing and computer vision driven by data and knowledge. We are seeking original research articles, reviews, and survey papers that explore the latest developments, challenges, and solutions in these rapidly advancing fields. Topics may include, but are not limited to, the following:
· Multimodal artificial intelligence
· 2D&3D generative modes
· Image & video segmentation
· 3D reconstruction
· Large models and their applications in signal processing and computer vision
· Visual question and answer (VQA), visual reasoning
· Meta-learning, transfer learning, few-shot learning.
· Embodied Artificial Intelligence
· Reinforcement Learning
· Medical image processing
· Medical robot
· Vision Foundation Models
· Efficient and robust AI
· Object detection and recognition
· Remote sensing image analysis
· Earth observation and geospatial intelligence
· Regional studies and environmental monitoring
· Self-supervised and semi-supervised learning for signal processing and visual tasks
Login
Register
Submit a Paper
Propose a Special lssue