Thousands of images and videos are collected from construction projects during construction. These contain valuable data that, if harnessed efficiently, can help automate or at least reduce human effort in diverse construction management activities such as progress monitoring, safety management, quality control and productivity tracking. Extracting meaningful information from images requires the development of technology and algorithms that enable computers to understand digital images or videos, replicating the functionality of human visual systems. This is the goal of computer vision. This review aims at providing an updated and categorized overview of computer vision applications in construction by examining the recent developments in the field and identifying the opportunities and challenges that future research needs to address to fully leverage the potential benefits of Computer Vision. We restrict the focus to four areas that can benefit the most from computer vision - Safety Management, Progress Monitoring, Productivity Tracking and Quality Control.

在施工过程中从建筑项目中收集了数以千计的图像和视频。这些包含有价值的数据，如果有效利用这些数据，可以帮助自动化或至少减少各种施工管理活动中的人力，例如进度监控、安全管理、质量控制和生产力跟踪。从图像中提取有意义的信息需要开发技术和算法，使计算机能够理解数字图像或视频，复制人类视觉系统的功能。这就是计算机视觉的目标。本综述旨在通过检查该领域的最新发展并确定未来研究需要解决的机遇和挑战，以充分利用计算机视觉的潜在优势，从而提供建筑中计算机视觉应用的更新和分类概述。我们将重点限制在可以从计算机视觉中受益最多的四个领域 - 安全管理、进度监控、生产力跟踪和质量控制。