Embedded Files
New decoder Results
New decoder Results
Max-Pooling
Max-Pooling
This image is reconstructed using vanilla WCT that uses max-pooling/unpooling in their encoder/decoders.
LIP
LIP
This image is reconstructed using encoder/decoders that uses LIP/unpooling. As we see the details are more clear, sharp, and the image has better contrast.
SurveY Results
SurveY Results
Results from our viewer survey
WCT
WCT
WCT was designed for artistic stylizations and did not take photo realism into consideration and hence suffers from structural artifacts and produces distortions on object boundaries. This can be clearly seen from the below stylized images produced by this technique. It was evident that the output images require some kind of smoothing step to make it photorealistic.
Comparison of WCT and our Approaches
Comparison of WCT and our Approaches
Content
style
Original WCT
PhotoWCT with incorporated smoothing
WCT2 with our segmentation maps
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
Original WCT
PhotoWCT with incorporated smoothing
WCT2 with our segmentation maps
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
Original WCT
PhotoWCT with incorporated smoothing
WCT2 with our segmentation maps
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
Original WCT
PhotoWCT with incorporated smoothing
WCT2 with our segmentation maps
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
Original WCT
PhotoWCT with incorporated smoothing
WCT2 with our segmentation maps
WCT2 using our segmentation maps
Our Approach: LIP + SPN
More Results
More Results
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Content
style
WCT2 using our segmentation maps
Our Approach: LIP + SPN
Page updated
Google Sites
Report abuse