Semantics Disentangling for Text-To-Image Generation
Authors:
- 1 June 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2322-2331
Abstract
Synthesizing photo-realistic images from text descriptions is a challenging problem. Previous studies have shown remarkable progresses on visual quality of the generated images. In this paper, we consider semantics from the input text descriptions in helping render photo-realistic images. However, diverse linguistic expressions pose challenges in extracting consistent semantics even they depict the same thing. To this end, we propose a novel photo-realistic text-to-image generation model that implicitly disentangles semantics to both fulfill the high-level semantic consistency and low-level semantic diversity. To be specific, we design (1) a Siamese mechanism in the discriminator to learn consistent high-level semantics, and (2) a visual-semantic embedding strategy by semantic-conditioned batch normalization to find diverse low-level semantics. Extensive experiments and ablation studies on CUB and MS-COCO datasets demonstrate the superiority of the proposed method in comparison to state-of-the-art methods.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Shadow Detection with Conditional Generative Adversarial Networks2017
- A Two Stream Siamese Convolutional Neural Network for Person Re-identification2017
- Plug & Play Generative Networks: Conditional Iterative Generation of Images in Latent Space2017
- Perceptual Generative Adversarial Networks for Small Object Detection2017
- Generative Attribute Controller with Conditional Filtered Generative Adversarial Networks2017
- Stacked Generative Adversarial Networks2017
- Multi-level Attention Networks for Visual Question Answering2017
- Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks2017
- Dimensionality Reduction by Learning an Invariant Mapping2006
- Long Short-Term MemoryNeural Computation, 1997