基于监督学习的单幅图像深度估计综述-计算机辅助设计与图形学学报.PDFVIP

30 8 Vol.30 No.8 2018 8 Journal of Computer-Aided Design Computer Graphics Aug. 2018 基于监督学习的单幅图像深度估计综述 * 毕天腾, 刘 越 , 翁冬冬, 王涌天 (北京理工大学光电学院 北京 100081) (liuyue@) 摘 要: 单幅图像深度估计是三维重建中基于图像获取场景深度的重要技术, 也是计算机视觉中的经典问题, 近年 来, 基于监督学习的单幅图像深度估计发展迅速. 文中介绍了基于监督学习的单幅图像深度估计及其模型和优化方 法; 分析了现有的参数学习、非参数学习、深度学习3 类方法及每类方法的国内外研究现状及优缺点; 最后对基于监 督学习的单幅图像深度估计进行总结, 得出了深度学习框架下的单幅图像深度估计是未来研究的发展趋势和重点. 关键词: 机器学习; 深度估计; 三维重建; 深度学习 中图法分类号: TP391.41 DOI: 10.3724/SP.J.1089.2018.16882 Survey on Supervised Learning Based Depth Estimation from a Single Image * Bi Tianteng, Liu Yue , Weng Dongdong, and Wang Yongtian (School of Optoelectronics, Beijing Institute of Technology, Beijing 100081) Abstract: Depth estimation from a single image is an important technology in the image-based depth acqui- sition for 3D reconstruction, which is also a classical problem in computer vision. Recently, supervised learning based depth estimation from a single image develops rapidly. In this paper, the recent related litera- tures are reviewed and supervised learning based depth estimation from a single image and its model and optimization are introduced. The current research situations of the parametric learning method, non-parametric learning method and deep learning method both in domestic and abroad are analyzed respectively with their advantages and disadvantages. At last, summarizing these methods leads to the conclusion that depth esti- mation from a single image in deep learning framework is the development trend and research priority in the future. Key words: machine learning; depth estimation; 3D reconstruction; deep learning [2-3].