TY - JOUR
T1 - Unsupervised Generation of Free-Form and Parameterized Avatars
AU - Polyak, Adam
AU - Taigman, Yaniv
AU - Wolf, Lior
N1 - Publisher Copyright:
© 1979-2012 IEEE.
PY - 2020/2/1
Y1 - 2020/2/1
N2 - We study two problems involving the task of mapping images between different domains. The first problem, transfers an image in one domain to an analog image in another domain. The second problem, extends the previous one by mapping an input image to a tied pair, consisting of a vector of parameters and an image that is created using a graphical engine from this vector of parameters. Similar to the first problem, the mapping's objective is to have the output image as similar as possible to the input image. In both cases, no supervision is given during training in the form of matching inputs and outputs. We compare the two unsupervised learning problems to the problem of unsupervised domain adaptation, define generalization bounds that are based on discrepancy, and employ a GAN to implement network solutions that correspond to these bounds. Experimentally, our methods are shown to solve the problem of automatically creating avatars.
AB - We study two problems involving the task of mapping images between different domains. The first problem, transfers an image in one domain to an analog image in another domain. The second problem, extends the previous one by mapping an input image to a tied pair, consisting of a vector of parameters and an image that is created using a graphical engine from this vector of parameters. Similar to the first problem, the mapping's objective is to have the output image as similar as possible to the input image. In both cases, no supervision is given during training in the form of matching inputs and outputs. We compare the two unsupervised learning problems to the problem of unsupervised domain adaptation, define generalization bounds that are based on discrepancy, and employ a GAN to implement network solutions that correspond to these bounds. Experimentally, our methods are shown to solve the problem of automatically creating avatars.
KW - Deep learning
KW - analysis by synthesis
KW - cross-domain transfer
KW - domain adaptation
KW - domain transfer network
KW - neural network
KW - tied output synthesis
UR - http://www.scopus.com/inward/record.url?scp=85051016719&partnerID=8YFLogxK
U2 - 10.1109/TPAMI.2018.2863282
DO - 10.1109/TPAMI.2018.2863282
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
C2 - 30080143
AN - SCOPUS:85051016719
SN - 0162-8828
VL - 42
SP - 444
EP - 459
JO - IEEE Transactions on Pattern Analysis and Machine Intelligence
JF - IEEE Transactions on Pattern Analysis and Machine Intelligence
IS - 2
M1 - 8425579
ER -