TY - GEN
T1 - Collapse resistant deep convolutional GAN for multi-object image generation
AU - Bolluyt, Elijah
AU - Comaniciu, Cristina
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/12
Y1 - 2019/12
N2 - This work introduces a novel system for the generation of images that contain multiple classes of objects. Recent work in Generative Adversarial Networks have produced high quality images, but many focus on generating images of a single object or set of objects. Our system addresses the task of image generation conditioned on a list of desired classes to be included in a single image. This enables our system to generate images with any given combination of objects, all composed into a visually realistic natural image. The system learns the interrelationships of all classes represented in a dataset, and can generate diverse samples including a set of these classes. It displays the ability to arrange these objects together, accounting for occlusions and inter-object spatial relations that characterize complex natural images. To accomplish this, we introduce a novel architecture based on Conditional Deep Convolutional GANs that is stabilized against collapse relative to both mode and condition. The system learns to rectify mode collapse during training, self-correcting to avoid suboptimal generation modes.
AB - This work introduces a novel system for the generation of images that contain multiple classes of objects. Recent work in Generative Adversarial Networks have produced high quality images, but many focus on generating images of a single object or set of objects. Our system addresses the task of image generation conditioned on a list of desired classes to be included in a single image. This enables our system to generate images with any given combination of objects, all composed into a visually realistic natural image. The system learns the interrelationships of all classes represented in a dataset, and can generate diverse samples including a set of these classes. It displays the ability to arrange these objects together, accounting for occlusions and inter-object spatial relations that characterize complex natural images. To accomplish this, we introduce a novel architecture based on Conditional Deep Convolutional GANs that is stabilized against collapse relative to both mode and condition. The system learns to rectify mode collapse during training, self-correcting to avoid suboptimal generation modes.
KW - Conditional Generation
KW - Generative Adversarial Networks
KW - Image Generation
KW - Minibatch Discrimination
KW - Multi-class generation
KW - Multi-class images
KW - Multi-object Images
KW - Multi-object generation
UR - http://www.scopus.com/inward/record.url?scp=85080910806&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85080910806&partnerID=8YFLogxK
U2 - 10.1109/ICMLA.2019.00229
DO - 10.1109/ICMLA.2019.00229
M3 - Conference contribution
AN - SCOPUS:85080910806
T3 - Proceedings - 18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019
SP - 1404
EP - 1408
BT - Proceedings - 18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019
A2 - Wani, M. Arif
A2 - Khoshgoftaar, Taghi M.
A2 - Wang, Dingding
A2 - Wang, Huanjing
A2 - Seliya, Naeem
T2 - 18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019
Y2 - 16 December 2019 through 19 December 2019
ER -