TY - GEN
T1 - Distilling Reflection Dynamics for Single-Image Reflection Removal
AU - Zheng, Quanlong
AU - Qiao, Xiaotian
AU - Cao, Ying
AU - Guo, Shi
AU - Zhang, Lei
AU - Lau, Rynson
N1 - Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).
PY - 2021
Y1 - 2021
N2 - Single-image reflection removal (SIRR) aims to restore the transmitted image given a single image shot through glass or window. Existing methods rely mainly on information extracted from a single image along with some predefined priors, and fail to give satisfying results on real-world images, due to inherent ambiguity and lack of large and diverse real-world training data. In this paper, instead of reasoning about a single image only, we propose to distill a representation of reflection dynamics from multi-view images (i.e., the motions of reflection and transmission layers over time), and transfer the learned knowledge for the SIRR problem. In particular, we propose a teacher-student framework where the teacher network learns a representation of reflection dynamics by watching a sequence of multi-view images of a scene captured by a moving camera and teaches a student network to remove reflection from a single input image. In addition, we collect a large real-world multi-view reflection image dataset for reflection dynamics knowledge distillation. Extensive experiments show that our model yields state-of-the-art performances.
AB - Single-image reflection removal (SIRR) aims to restore the transmitted image given a single image shot through glass or window. Existing methods rely mainly on information extracted from a single image along with some predefined priors, and fail to give satisfying results on real-world images, due to inherent ambiguity and lack of large and diverse real-world training data. In this paper, instead of reasoning about a single image only, we propose to distill a representation of reflection dynamics from multi-view images (i.e., the motions of reflection and transmission layers over time), and transfer the learned knowledge for the SIRR problem. In particular, we propose a teacher-student framework where the teacher network learns a representation of reflection dynamics by watching a sequence of multi-view images of a scene captured by a moving camera and teaches a student network to remove reflection from a single input image. In addition, we collect a large real-world multi-view reflection image dataset for reflection dynamics knowledge distillation. Extensive experiments show that our model yields state-of-the-art performances.
UR - https://www.scopus.com/pages/publications/85123056445
UR - https://www.scopus.com/pages/publications/85123056445#tab=impact
U2 - 10.1109/ICCVW54120.2021.00215
DO - 10.1109/ICCVW54120.2021.00215
M3 - RGC 32 - Refereed conference paper (with host publication)
SN - 978-1-6654-0192-0
VL - 2021-October
T3 - Proceedings of the IEEE International Conference on Computer Vision Workshops
SP - 1886
EP - 1894
BT - Proceedings - 2021 IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2021
PB - IEEE
T2 - 18th IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2021
Y2 - 11 October 2021 through 17 October 2021
ER -