Generating Annotated High-Fidelity Images Containing Multiple Coherent Objects

TL;DR

This paper introduces a novel multi-object image generation framework that produces high-fidelity, coherent images with object annotations without needing explicit contextual information, useful for domains like medical imaging.

Contribution

The proposed method combines VQ-VAE with autoregressive priors PixelSNAIL and LayoutPixelSNAIL to generate multi-object images with preserved spatial and semantic coherence without auxiliary input.

Findings

Outperforms state-of-the-art multi-object generative methods on Multi-MNIST and CLEVR datasets.

Generated images maintain high fidelity and object coherence.

Augmenting training data with generated images improves model performance in medical imaging.

Abstract

Recent developments related to generative models have made it possible to generate diverse high-fidelity images. In particular, layout-to-image generation models have gained significant attention due to their capability to generate realistic complex images containing distinct objects. These models are generally conditioned on either semantic layouts or textual descriptions. However, unlike natural images, providing auxiliary information can be extremely hard in domains such as biomedical imaging and remote sensing. In this work, we propose a multi-object generation framework that can synthesize images with multiple objects without explicitly requiring their contextual information during the generation process. Based on a vector-quantized variational autoencoder (VQ-VAE) backbone, our model learns to preserve spatial coherency within an image as well as semantic coherency between the objects and the background through two powerful autoregressive priors: PixelSNAIL and LayoutPixelSNAIL. While the PixelSNAIL learns the distribution of the latent encodings of the VQ-VAE, the LayoutPixelSNAIL is used to specifically learn the semantic distribution of the objects. An implicit advantage of our approach is that the generated samples are accompanied by object-level annotations. We demonstrate how coherency and fidelity are preserved with our method through experiments on the Multi-MNIST and CLEVR datasets; thereby outperforming state-of-the-art multi-object generative methods. The efficacy of our approach is demonstrated through application on medical imaging datasets, where we show that augmenting the training set with generated samples using our approach improves the performance of existing models.