Machine Vision and Neural Computing 机器视觉与神经计算

Introduction:

The combination of machine vision and neural computing is an important research direction in the field of artificial intelligence, and has achieved remarkable progress in recent years. With the rapid development of deep learning technology, models such as convolutional neural networks (CNN), recurrent neural networks (RNN), and Transformer have performed well in tasks such as image recognition, object detection, and semantic segmentation. Emerging technologies such as multimodal learning and self-supervised learning have further enhanced the intelligence level of machine vision, enabling it to learn from and adapt to complex scenarios with a small amount of labeled data. In terms of hardware, the popularity of high-performance computing devices such as GPUs and TPUs has provided strong computing power support for neural computing, facilitating the realization of real-time visual processing and large-scale model training.
Although the combination of machine vision and neural computing has achieved significant results, it still faces many challenges. The training of deep learning models requires a large amount of labeled data, and the cost of data acquisition and labeling is high, and there are privacy issues in some fields. The interpretability of existing models is poor, making it difficult to meet the requirements of high-reliability applications. The high energy consumption and computational complexity of neural computing limit its deployment on edge devices. Models are vulnerable when facing unknown scenarios or adversarial samples, and have insufficient generalization ability. These problems urgently require joint efforts from academia and industry to be solved through technological innovation and interdisciplinary collaboration.
This topic focuses on the frontier theories and engineering practices of machine vision and neural computing, aiming to exchange the latest research results in related fields. The topics include, but are not limited to: image recognition and segmentation based on deep learning, multimodal perception and fusion, lightweight neural network design, self-supervised and unsupervised learning methods, research on model interpretability and robustness, and innovative applications of machine vision in autonomous driving, medical care, industry, etc. We look forward to your participation and jointly promoting the frontier development of machine vision and neural computing.

机器视觉与神经计算的结合是人工智能领域的重要研究方向，近年来取得了显著进展。随着深度学习技术的快速发展，卷积神经网络（CNN）、循环神经网络（RNN）和Transformer等模型在图像识别、目标检测、语义分割等任务中表现出色。多模态学习和自监督学习等新兴技术进一步提升了机器视觉的智能化水平，使其能够从少量标注数据中学习并适应复杂场景。硬件方面，GPU、TPU等高性能计算设备的普及也为神经计算提供了强大的算力支持，推动了实时视觉处理和大规模模型训练的实现。
尽管机器视觉与神经计算的结合取得了显著成果，但仍面临诸多挑战。深度学习模型的训练需要大量标注数据，而数据获取和标注成本高昂，且在某些领域存在隐私问题。现有模型的可解释性较差，难以满足高可靠性应用的需求。神经计算的高能耗和计算复杂度限制了其在边缘设备上的部署。模型在面对未知场景或对抗样本时表现脆弱，泛化能力不足。这些问题亟待学术界和工业界共同努力，通过技术创新和跨学科合作加以解决。
本专题聚焦机器视觉与神经计算的前沿理论与工程实践，旨在交流相关领域的最新研究成果，主题包括但不限于：基于深度学习的图像识别与分割、多模态感知与融合、轻量化神经网络设计、自监督与无监督学习方法、模型可解释性与鲁棒性研究、以及机器视觉在自动驾驶、医疗、工业等领域的创新应用。我们期待您的参与，共同推动机器视觉与神经计算的前沿发展。

Organizers:

	Yaping Yang, Xihang University, China Yaping Yang is the dean and Professor of the school of electronic engineering at Xihang University. She has been committed to the research of measurement and control technology and automation system control for a long time, and has led and participated in 5 provincial-level scientific research projects, 4 department level scientific research projects, and more than 20 school level teaching and research projects. In recent years, she has been published more than 10 papers, and has won the second prize of the school level scientific research award and the third prize of the Qin Chuangyuan Fengdong Cup Shaanxi Province Science and Technology Worker Innovation and Entrepreneurship Competition.
	Xuejuan Kang, Xihang University, China Xuejuan Kang is the Associate Professor with the School of Electronic Engineering, Xihang University. With a long-standing focus on the research of artificial intelligence and industrial image surface anomaly detection, he has led and participated in numerous research projects, including three projects of Shaanxi Provincial Science and Technology Department and four projects of Shaanxi Provincial Education Department. In recent years, she has published many high-quality papers in international journals such as Textile Research Journal, IEEE Access, and Journal of Engineered Fibers and Fabrics. Her in-depth theoretical research and practical application in the visual detection of fabric surface defects under complex texture background, and has also earned several prestigious awards, including the second prize of Shaanxi Science and Technology Award, first and second prize of Shaanxi Higher Education Institutions Science and Technology Award.
	He Song, Xihang University, China He Song received the Ph.D. degrees in control theory and control engineering from the Xi'an University of Technology, Xi’an, China, in 2024. He is currently a lecturer with the School of Electronic Engineering, Xihang University. He has published more than 10 academic papers, and has participated in multiple national and provincial-level projects. His current research interests include intelligent control algorithms and its application to flexible systems, and navigation control based on deep learning.

Submission Guideline:

Please submit your manuscript via Online Submission System: https://easychair.org/conferences/?conf=rcae2025
Please choose Special Session: Machine Vision and Neural Computing