Learning to See Through Obstructions
Back
Results on reflection removal
Dataset:
[Xue et al. 2015]
[Li and Brown 2013]
[Guo et al. 2014]
[Kopf et al. 2013]
Frame:
01
/
10
: Toy
Method:
Input
[Fan et al. 2017]
[Zhang et al. 2018]
[Yang et al. 2018]
[Wei et al. 2019]
[Jin et al. 2018]
[Gandelsman et al. 2019]
[Sinha et al. 2012]
[Li and Brown 2013]
[Guo et al. 2014]
[Xue et al. 2015]
[Alayrac et al. 2019]
Ours
[Xue et al. 2015] dataset link
Remarks: Multiple frames as input.
Reconstructed background
Reconstructed reflection
Input sequence
