Glass Detection Dataset

Haiyang Mei,Xin Yang,Yang Wang,Yuanyuan LiuShengfeng He,Qiang Zhang,Xiaopeng Wei,Rynson W.H. Lau.

The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2020.

https://openaccess.thecvf.com/content_CVPR_2020/papers/Mei_Dont_Hit_Me_Glass_Detection_in_Real-World_Scenes_CVPR_2020_paper.pdf

Mei_Dont_Hit_Me_Glass_Detection_in_Real-World_Scenes_CVPR_2020_paper.pdf

Please cite this paper when using the data：BibTex.txt

Introduction:

Glass is very common in our daily life. Existing computer vision systems neglect the glass and thus might lead to

severe consequence,  e.g., the robot might crash into the glass wall. However, sensing the presence of the glass

is not straightforward. The key challenge is that arbitrary objects/scenes can appear behind the glass and the co-

ntent presented in the glass region typically similar to those outside of it. In this paper, we raise an interesting but 

important problem of detecting glass from a single RGB image. To address this problem, we construct a large-sc-

ale glass detection dataset (GDD) and design a glass detection network, called GDNet, by learning abundant con-

textual features from a global perspective with a novel large-field contextual feature integration module. Extensive

experiments demonstrate the proposed method achieves superior glass detection results on our GDD test set. Pa-

rticularly, we outperform state-of-the-art methods that fine-tuned for glass detection.