THE 3
rd
INTERNATIONAL SCIENTIFIC CONFERENCES OF STUDENTS AND YOUNG RESEARCHERS
dedicated to the 99
th
anniversary of the National Leader
of Azerbaijan Heydar Aliyev
159
Alphabet. Cybernetics and Computer Technologies. 2021. 3. P. 65–73.
https://doi.org/10.34229/2707-451X.21.3.6
[2] N. Kato, M. Suzuki, S. Omachi, H. Aso and Y. Nemoto, "A handwritten character
recognition system using directional element feature and asymmetric Mahalanobis
distance," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21, no.
3, pp. 258-262, March 1999, doi: 10.1109/34.754617.
[3] Lecun, Yann & Bottou, Leon & Bengio, Y. & Haffner, Patrick. (1998).
Gradient-Based
Learning Applied to Document Recognition. Proceedings of the IEEE. 86. 2278 - 2324.
10.1109/5.726791.
IMAGE GENERATION BASED ON KINSHIP ANALYSIS
Eyvaz Najafli
Baku Higher Oil School
Baku, Azerbaijan
eyvaz.necefli.std@bhos.edu.az
Supervisor: Ph.D Associate Professor Ali Parsayan
Keywords:
Generative Artificial Intelligence,
Image Generation, StyleGAN, Style
Encoding, Image Translation, Convolutional Neural Networks, Latent Feature Mapping, Deep
Learning
Modelled Kinship
Verification systems
which incorporate the automatic latent
facial
feature
extraction
and
the
comparison have gained enormous
interest
from
research
community.
Recently, AI scientists have endeavoured
to create systems
that can solve inverse
problem, in other words, generate possible
kin images from the given input images.
Many works in this area have focused on
usage of GANs (Generative Adversarial
Networks) in order to recreate kin faces
from transformed latent features. In this
paper, I will
present a new approach to
generate possible child images by
mapping parent images into decoupled
latent space
and decoding the result via
StyleGAN Generator which produces more linear and less entangled
representation [1].
Dostları ilə paylaş: