Matches in SemOpenAlex for { <https://semopenalex.org/work/W2901051804> ?p ?o ?g. }
- W2901051804 abstract "Unsupervised domain mapping aims to learn a function to translate domain X to Y by a function GXY in the absence of paired examples. Finding the optimal GXY without paired data is an ill-posed problem, so appropriate constraints are required to obtain reasonable solutions. One of the most prominent constraints is cycle consistency, which enforces the translated image by GXY to be translated back to the input image by an inverse mapping GYX. While cycle consistency requires the simultaneous training of GXY and GY X, recent studies have shown that one-sided domain mapping can be achieved by preserving pairwise distances between images. Although cycle consistency and distance preservation successfully constrain the solution space, they overlook the special properties that simple geometric transformations do not change the semantic structure of images. Based on this special property, we develop a geometry-consistent generative adversarial network (GcGAN), which enables one-sided unsupervised domain mapping. GcGAN takes the original image and its counterpart image transformed by a predefined geometric transformation as inputs and generates two images in the new domain coupled with the corresponding geometry-consistency constraint. The geometry-consistency constraint reduces the space of possible solutions while keep the correct solutions in the search space. Quantitative and qualitative comparisons with the baseline (GAN alone) and the state-of-the-art methods including CycleGAN and DistanceGAN demonstrate the effectiveness of our method." @default.
- W2901051804 created "2018-11-29" @default.
- W2901051804 creator A5010516895 @default.
- W2901051804 creator A5012429599 @default.
- W2901051804 creator A5016038041 @default.
- W2901051804 creator A5031717623 @default.
- W2901051804 creator A5074103823 @default.
- W2901051804 creator A5091227928 @default.
- W2901051804 date "2018-09-16" @default.
- W2901051804 modified "2023-09-23" @default.
- W2901051804 title "Geometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain Mapping" @default.
- W2901051804 cites W1522301498 @default.
- W2901051804 cites W1903029394 @default.
- W2901051804 cites W2083366168 @default.
- W2901051804 cites W2099471712 @default.
- W2901051804 cites W2108598243 @default.
- W2901051804 cites W2194775991 @default.
- W2901051804 cites W2295130376 @default.
- W2901051804 cites W2331128040 @default.
- W2901051804 cites W2339754110 @default.
- W2901051804 cites W2342877626 @default.
- W2901051804 cites W2411541852 @default.
- W2901051804 cites W2412320034 @default.
- W2901051804 cites W2423557781 @default.
- W2901051804 cites W2431874326 @default.
- W2901051804 cites W2432004435 @default.
- W2901051804 cites W2502312327 @default.
- W2901051804 cites W2553897675 @default.
- W2901051804 cites W2564591810 @default.
- W2901051804 cites W2592480533 @default.
- W2901051804 cites W2593414223 @default.
- W2901051804 cites W2608015370 @default.
- W2901051804 cites W2623767209 @default.
- W2901051804 cites W2737258237 @default.
- W2901051804 cites W2739540493 @default.
- W2901051804 cites W2768368415 @default.
- W2901051804 cites W2768959015 @default.
- W2901051804 cites W2770646061 @default.
- W2901051804 cites W2788945792 @default.
- W2901051804 cites W2796286778 @default.
- W2901051804 cites W2798454267 @default.
- W2901051804 cites W2798600195 @default.
- W2901051804 cites W2799062770 @default.
- W2901051804 cites W2807033398 @default.
- W2901051804 cites W2883376126 @default.
- W2901051804 cites W2887513558 @default.
- W2901051804 cites W2949272014 @default.
- W2901051804 cites W2949999304 @default.
- W2901051804 cites W2951523806 @default.
- W2901051804 cites W2951939904 @default.
- W2901051804 cites W2953127297 @default.
- W2901051804 cites W2953139137 @default.
- W2901051804 cites W2962793481 @default.
- W2901051804 cites W2962808524 @default.
- W2901051804 cites W2962851930 @default.
- W2901051804 cites W2962968458 @default.
- W2901051804 cites W2963047159 @default.
- W2901051804 cites W2963073614 @default.
- W2901051804 cites W2963344645 @default.
- W2901051804 cites W2963470893 @default.
- W2901051804 cites W2963684088 @default.
- W2901051804 cites W2963709863 @default.
- W2901051804 cites W2963784072 @default.
- W2901051804 cites W2963801463 @default.
- W2901051804 cites W2963890275 @default.
- W2901051804 cites W2964168187 @default.
- W2901051804 cites W2964271895 @default.
- W2901051804 cites W3159890710 @default.
- W2901051804 cites W2798873573 @default.
- W2901051804 doi "https://doi.org/10.48550/arxiv.1809.05852" @default.
- W2901051804 hasPublicationYear "2018" @default.
- W2901051804 type Work @default.
- W2901051804 sameAs 2901051804 @default.
- W2901051804 citedByCount "1" @default.
- W2901051804 countsByYear W29010518042018 @default.
- W2901051804 crossrefType "posted-content" @default.
- W2901051804 hasAuthorship W2901051804A5010516895 @default.
- W2901051804 hasAuthorship W2901051804A5012429599 @default.
- W2901051804 hasAuthorship W2901051804A5016038041 @default.
- W2901051804 hasAuthorship W2901051804A5031717623 @default.
- W2901051804 hasAuthorship W2901051804A5074103823 @default.
- W2901051804 hasAuthorship W2901051804A5091227928 @default.
- W2901051804 hasBestOaLocation W29010518041 @default.
- W2901051804 hasConcept C104317684 @default.
- W2901051804 hasConcept C111472728 @default.
- W2901051804 hasConcept C11413529 @default.
- W2901051804 hasConcept C115961682 @default.
- W2901051804 hasConcept C134306372 @default.
- W2901051804 hasConcept C137105694 @default.
- W2901051804 hasConcept C138885662 @default.
- W2901051804 hasConcept C14036430 @default.
- W2901051804 hasConcept C154945302 @default.
- W2901051804 hasConcept C184898388 @default.
- W2901051804 hasConcept C185592680 @default.
- W2901051804 hasConcept C189950617 @default.
- W2901051804 hasConcept C204241405 @default.
- W2901051804 hasConcept C2524010 @default.
- W2901051804 hasConcept C2776036281 @default.
- W2901051804 hasConcept C2776436953 @default.
- W2901051804 hasConcept C33923547 @default.