Realistic Image Synthesis with Configurable 3D Scene Layouts

TL;DR

This paper introduces a novel method for realistic image synthesis from 3D scene layouts, enabling precise control over object positions, orientations, and styles while maintaining geometric correctness.

Contribution

It proposes a 3D scene painting network trained via 2D semantic synthesis to generate realistic images with controllable geometry and style, addressing limitations of previous methods.

Findings

Produces images with geometrically valid structures

Supports viewpoint and object pose manipulation

Enables style manipulation of synthesized images

Abstract

Recent conditional image synthesis approaches provide high-quality synthesized images. However, it is still challenging to accurately adjust image contents such as the positions and orientations of objects, and synthesized images often have geometrically invalid contents. To provide users with rich controllability on synthesized images in the aspect of 3D geometry, we propose a novel approach to realistic-looking image synthesis based on a configurable 3D scene layout. Our approach takes a 3D scene with semantic class labels as input and trains a 3D scene painting network that synthesizes color values for the input 3D scene. With the trained painting network, realistic-looking images for the input 3D scene can be rendered and manipulated. To train the painting network without 3D color supervision, we exploit an off-the-shelf 2D semantic image synthesis method. In experiments, we show that our approach produces images with geometrically correct structures and supports geometric manipulation such as the change of the viewpoint and object poses as well as manipulation of the painting style.