Learning to Parse Wireframes in Images of Man-Made Environments

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.

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract


Introduced in the Paper:


Used in the Paper:

York Urban Line Segment Database
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Line Segment Detection wireframe dataset Wireframe parser F1 score 0.728 # 6
sAP5 3.7 # 10
sAP10 5.1 # 12
sAP15 5.9 # 7
Line Segment Detection York Urban Dataset Wireframe parser F1 score 0.627 # 7
sAP5 2.8 # 10
sAP10 3.9 # 12

Results from Other Papers

Task Dataset Model Metric Name Metric Value Rank Uses Extra
Training Data
Source Paper Compare
Line Segment Detection wireframe dataset WF-Parser sAP5 6.9 # 9
sAP10 9.0 # 11


No methods listed for this paper. Add relevant methods here