Unsupervised Visual Attribute Transfer with Reconfigurable Generative Adversarial Networks

By Taeksoo Kim, Byoungjip Kim, Moonsu Cha, Jiwon Kim. 2017

42 citations recorded by the source corpus.

Learning to transfer visual attributes requires supervision dataset. Corresponding images with varying attribute values with the same identity are required for learning the transfer function. This largely limits their applications, because capturing them is often a difficult task. To address the issue, we propose an unsupervised method to learn to transfer visual attribute. The proposed method can learn the transfer function without any corresponding images. Inspecting visualization results from various unsupervised attribute transfer tasks, we verify the effectiveness of the proposed method.

Subfields: Computer Vision and Pattern Recognition, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Sungroh Yoon
2. Yongjun Hong
3. Shiguo Lian
4. Kai Wang
5. Sunghun Kim
6. J. Choo
7. Munyoung Kim
8. Xiang Wang

In Generative Models

1. Shiguo Lian
2. Sunghun Kim
3. J. Choo
4. Munyoung Kim
5. Yongjun Hong
6. Xiang Wang
7. Kai Wang
8. Jung-Woo Ha

Citations in the corpus

• StarGAN: Unified Generative Adversarial Networks for Multi-domain Image-to-Image Translation (2017, background): “Generative adversarial networks (GANs) [3] have shown remarkable results in various computer vision tasks such as image generation [1, 6, 23, 31], image translation [7, 8, 32], super-resolution imaging [13], and face image synthesis [9, 15, 25, 30].”
• How Generative Adversarial Networks and Its Variants Work : An Overview of GAN (2017, background): “[57] decompose an attribute vector and aim to only transfer target attributes while non-target attributes are kept the same.”
• A survey on face data augmentation for the training of deep neural networks (2019, methodology): “Besides the color, [67] proposed an unsupervised visual attribute transfer using reconfigurable generative adversarial net-”
• How Generative Adversarial Networks and Their Variants Work (2017, background): “Kim et al. [58] attempted to transfer visual attributes.”