Unsupervised Visual Attribute Transfer with Reconfigurable Generative Adversarial Networks
This addresses a limitation in computer vision by enabling attribute transfer without costly labeled data, though it appears incremental as it builds on existing GAN frameworks.
The paper tackles the problem of visual attribute transfer without requiring supervised datasets of corresponding images, proposing an unsupervised method that learns the transfer function effectively across various tasks.
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.