AI-Driven Robot Vision: Progress, Challenges, and Perspectives

Dear Colleagues,

Robot vision serves as a fundamental technology enabling intelligent robots to perceive, understand, and interact with their environments. In recent years, the rapid evolution of artificial intelligence, particularly the rise of Convolutional Neural Networks (CNNs) and Vision Transformers, has significantly enhanced the precision, speed, and efficiency of visual understanding systems. These advancements have driven remarkable progress in core tasks such as semantic segmentation, 3D machine vision, and complex pattern recognition.

Furthermore, the integration of multimodal data (e.g., RGB, depth, point clouds, and tactile information) and the rapid development of Embodied AI have empowered robots with unprecedented capabilities. Modern visual systems can now achieve robust object detection and tracking, comprehensive scene understanding, and precise robotic grasping across highly diverse and uncontrolled environments.

Despite these remarkable breakthroughs, the field continues to face critical technical challenges. Persistent issues include ensuring robustness in dynamic environments, handling severe occlusions or perceiving challenging targets, and achieving real-time processing efficiency.

This Special Issue focuses on cutting-edge research in robot vision, covering key areas including visual SLAM, object detection and tracking, scene understanding, visual servoing, and human–robot interaction. We also address technical challenges in dynamic environments, occlusion handling, and real-time processing requirements. Original research on novel sensor fusion, end-to-end learning, neural radiance fields applications, and embodied AI vision are particularly welcome. This issue aims to promote innovative applications of robot vision in industrial manufacturing, autonomous driving, medical surgery, and service robotics, while exploring future development trends.

Research areas may include, but are not limited to, the following:

• Image and video analysis, pattern recognition, and semantic segmentation in complex scenarios;
• Three-dimensional machine vision, point cloud registration, and processing;
• Self-supervised, semi-supervised, and few-shot learning;
• Generative models and foundation models for vision tasks;
• Pose estimation;
• Depth estimation;
• Object detection, tracking, and perception of challenging targets;
• Visual SLAM, scene reconstruction, and environment understanding;
• Vision-driven robotic grasping, manipulation, and visual servoing;
• Multimodal and cross-domain representation learning for robot vision;
• Embodied AI and human–robot interaction.

Ultimately, this Special Issue seeks to bridge the critical gap between advanced visual perception and autonomous robotic action. By assembling these diverse research efforts, we aim to pave the way for robust, adaptable, and highly intelligent robotic systems capable of thriving in unconstrained real-world environments.

Dr. Deping Li
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. 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.

• robot vision
• deep learning
• vision transformer
• 3D perception and point cloud
• generative models
• visual SLAM
• 3D perception
• pose estimation
• depth estimation
• object tracking
• visual servoing
• multi-modal fusion
• embodied AI
• autonomous driving