Applications of Computer Vision for Autonomous Driving

Dear Colleagues,

Computer vision is a cornerstone of autonomous driving, enabling vehicles to perceive, understand, and anticipate complex, dynamic environments. Despite rapid progress in 2D/3D perception and multi-sensor fusion, deploying reliable and efficient vision systems at scale remains challenging due to long-tail events, adverse weather and illumination, domain shifts, compute and memory constraints on embedded hardware, and stringent safety and validation requirements. This Special Issue aims to gather advances that address these challenges and translate cutting-edge research into robust, real-world autonomy.

We welcome contributions spanning the full perception stack—from sensing and calibration to on-vehicle deployment and evaluation. Topics of interest include camera-, LiDAR-, and radar-based perception; 2D/3D object detection and tracking; semantic and instance segmentation; depth estimation; BEV scene representations and occupancy prediction; and tightly coupled multi-modal fusion. We particularly encourage work that quantifies and mitigates uncertainty, improves calibration and spatiotemporal alignment, and leverages priors for structure and motion understanding in complex traffic scenes.

Learning paradigms that advance data efficiency and generalization are in scope, including self-supervised and semi-supervised learning, curriculum and continual learning, open-vocabulary and foundation models for driving, and diffusion/generative models for augmentation and simulation. We welcome sim-to-real and domain adaptation methods that bridge gaps across sensors, geographies, weather, and camera rigs, as well as techniques for rare-event discovery, scenario mining, and active learning that target long-tail safety-critical behaviors.

Resource-aware vision for production autonomy is also a central theme: real-time and energy-efficient models; quantization, pruning, and neural architecture/search for automotive SoCs; compiler and runtime co-design; and robust performance under compute, latency, and memory constraints. Cooperative and connected intelligence—including V2X-enabled cooperative perception and map priors—are within scope, as are privacy-preserving learning and secure perception.

Finally, we seek rigorous evaluation, benchmarking, and verification contributions: protocols and datasets for robustness, reliability, and corner-case testing; uncertainty calibration and explainability; failure analysis and retraining pipelines; and reproducible systems with open data/code. Submissions from both academia and industry are encouraged, including case studies and system papers that demonstrate measurable gains on real-world fleets or large-scale public benchmarks.

This Special Issue invites original research articles, communications, and reviews that advance trustworthy, efficient, and deployable computer vision for autonomous and connected vehicles.

Dr. Wei Zhou
Dr. Weijie Yu
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 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-anonymized 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.

• autonomous driving
• computer vision
• multi-sensor fusion (camera–LiDAR–radar)
• 3D perception and BEV scene understanding
• robustness and uncertainty estimation
• domain adaptation and sim-to-real
• efficient on-vehicle deployment
• cooperative perception (V2X)