PasteGAN: A Semi-Parametric Method to Generate Image from Scene Graph

TL;DR

PasteGAN is a semi-parametric approach that generates controllable, high-quality images from scene graphs and object crops, allowing detailed manipulation of object appearances and interactions.

Contribution

The paper introduces PasteGAN, a novel semi-parametric method that combines scene graphs and object crops for more controllable and detailed image synthesis.

Findings

Outperforms SOTA on Inception Score, Diversity Score, FID

Generates complex and diverse images with specified objects

Effectively integrates object crops and scene graph relationships

Abstract

Despite some exciting progress on high-quality image generation from structured(scene graphs) or free-form(sentences) descriptions, most of them only guarantee the image-level semantical consistency, i.e. the generated image matching the semantic meaning of the description. They still lack the investigations on synthesizing the images in a more controllable way, like finely manipulating the visual appearance of every object. Therefore, to generate the images with preferred objects and rich interactions, we propose a semi-parametric method, PasteGAN, for generating the image from the scene graph and the image crops, where spatial arrangements of the objects and their pair-wise relationships are defined by the scene graph and the object appearances are determined by the given object crops. To enhance the interactions of the objects in the output, we design a Crop Refining Network and an Object-Image Fuser to embed the objects as well as their relationships into one map. Multiple losses work collaboratively to guarantee the generated images highly respecting the crops and complying with the scene graphs while maintaining excellent image quality. A crop selector is also proposed to pick the most-compatible crops from our external object tank by encoding the interactions around the objects in the scene graph if the crops are not provided. Evaluated on Visual Genome and COCO-Stuff dataset, our proposed method significantly outperforms the SOTA methods on Inception Score, Diversity Score and Fr\'echet Inception Distance. Extensive experiments also demonstrate our method's ability to generate complex and diverse images with given objects.