计算机科学
人工智能
计算机视觉
突出
解析
卷积神经网络
编码
对象(语法)
编码(集合论)
特征(语言学)
目标检测
任务(项目管理)
直线(几何图形)
代表(政治)
特征提取
模式识别(心理学)
源代码
几何学
生物化学
政治学
法学
程序设计语言
管理
数学
化学
集合(抽象数据类型)
经济
语言学
哲学
操作系统
基因
政治
作者
Kun Huang,Yifan Wang,Zihan Zhou,Tianjiao Ding,Shenghua Gao,Yi Ma
标识
DOI:10.1109/cvpr.2018.00072
摘要
In this paper, we propose a learning-based approach to the task of automatically extracting a "wireframe" representation for images of cluttered man-made environments. The wireframe (see Fig. 1) contains all salient straight lines and their junctions of the scene that encode efficiently and accurately large-scale geometry and object shapes. To this end, we have built a very large new dataset of over 5,000 images with wireframes thoroughly labelled by humans. We have proposed two convolutional neural networks that are suitable for extracting junctions and lines with large spatial support, respectively. The networks trained on our dataset have achieved significantly better performance than state-of-the-art methods for junction detection and line segment detection, respectively. We have conducted extensive experiments to evaluate quantitatively and qualitatively the wireframes obtained by our method, and have convincingly shown that effectively and efficiently parsing wireframes for images of man-made environments is a feasible goal within reach. Such wireframes could benefit many important visual tasks such as feature correspondence, 3D reconstruction, vision-based mapping, localization, and navigation. The data and source code are available at https://github.com/huangkuns/wireframe.
科研通智能强力驱动
Strongly Powered by AbleSci AI