follow-up tasks in autonomous driving[2], including 3D object recognition, visual navigation and positioning, and autonomous driving decision-making.Conventionally, depth information is obtained by depth sensors that involve LiDAR[3,4]or stereo cameras[5,6]. Nevertheless, the former is a prohibitively costly piece of equipment, and the depth maps obtained through dimensionality reduction are narrow in view and sparse in content. While the latter is more cost-effective, its accuracy in calculating parallax is constrained by structural concerns such as baseline length, and the stereo matching algorithms involved are computationally intensive[7,8].With the emergence of Convolutional Neural Networks (CNNs) technology [9-13], attempts have been undertaken to train network models capable of recovering depth information from monocular images. These methods notably minimize expenditures about equipment and computational resources. However, they introduce an unconventional challenge: training depth estimation models requires highprecision depth sensor measurements as ground-truth data. Additionally, collecting and annotating the image datasets is a labor-intensive and costly process.

Peng Guo

Shuguo Pan

Wang Gao

Kourosh Khoshelham

Machine Vision and Applications

机器视觉和应用程序发布机器视觉研究和开发方面的高质量技术贡献。具体来说，编辑鼓励在图像相关计算的所有应用和工程方面提交。特别是涉及机器视觉的科学、商业、工业、军事和生物医学应用的原创性贡献都在该期刊的范围内。特别强调图像处理和计算机视觉的工程和技术方面。以下方面机器视觉应用的兴趣：算法、架构、VLSI 实现、人工智能技术和机器视觉专家系统、前端传感、多维和多传感器机器视觉、实时技术、图像数据库、虚拟现实和可视化。论文必须包含重要的实验验证部分。

The journal "Machine Vision and Applications" publishes high-quality technical contributions in the field of machine vision research and development. Specifically, we encourage submissions related to all applications and engineering aspects of image-related computing. Original contributions involving scientific, commercial, industrial, military, and biomedical applications of machine vision are particularly welcome. The journal emphasizes the engineering and technical aspects of image processing and computer vision. We are especially interested in the following areas of machine vision applications: algorithms, architectures, VLSI implementations, artificial intelligence technologies and machine vision expert systems, front-end sensing, multi-dimensional and multi-sensor machine vision, real-time technologies, image databases, virtual reality, and visualization. All submitted papers must include significant experimental validation.

《Machine Vision and Applications》期刊致力于发布机器视觉领域的高质量技术研究和开发成果。我们鼓励提交与图像相关计算的各类应用和工程研究，尤其是涉及科学、商业、工业、军事和生物医学等领域的原创性贡献。期刊特别强调图像处理和计算机视觉的工程及技术方面。我们对以下机器视觉应用领域特别感兴趣：算法、架构、VLSI实现、人工智能技术和机器视觉专家系统、前端传感器、多维和多传感器机器视觉、实时技术、图像数据库、虚拟现实和可视化。所有提交的论文必须包含重要的实验验证部分。

Category	Quartile
计算机科学	4
计算机科学, 计算机人工智能	4
计算机科学, 计算机控制论	4
计算机科学, 工程电子与电气	4

Self-supervised monocular depth estimation via joint attention and intelligent mask loss

ivySCI AI Smartly Parses PDF, Answers Researchers' Questions, and Helps You Understand Papers in Seconds

Journal Info

Category	Quartile
ENGINEERING, ELECTRICAL & ELECTRONIC	2