Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network
This addresses the challenge of text-to-image synthesis for applications like content creation, with incremental advancements in network architecture and training objectives.
The paper tackles the problem of generating photographic images from text descriptions by introducing a hierarchically-nested adversarial network, resulting in significant improvements over previous state-of-the-art methods across multiple datasets and evaluation metrics.
This paper presents a novel method to deal with the challenging task of generating photographic images conditioned on semantic image descriptions. Our method introduces accompanying hierarchical-nested adversarial objectives inside the network hierarchies, which regularize mid-level representations and assist generator training to capture the complex image statistics. We present an extensile single-stream generator architecture to better adapt the jointed discriminators and push generated images up to high resolutions. We adopt a multi-purpose adversarial loss to encourage more effective image and text information usage in order to improve the semantic consistency and image fidelity simultaneously. Furthermore, we introduce a new visual-semantic similarity measure to evaluate the semantic consistency of generated images. With extensive experimental validation on three public datasets, our method significantly improves previous state of the arts on all datasets over different evaluation metrics.