REPARO: Compositional 3D Assets Generation with Differentiable 3D Layout Alignment

TL;DR

REPARO is a novel method for generating coherent 3D scenes with multiple objects from single images, using a two-step process of object extraction and differentiable layout optimization to improve scene realism.

Contribution

REPARO introduces a new compositional 3D asset generation framework that combines object reconstruction with differentiable layout optimization, addressing multi-object scene complexity.

Findings

Enhances object independence and scene coherence.

Improves detail accuracy in multi-object 3D scenes.

Demonstrates significant performance on multi-object scene generation.

Abstract

Traditional image-to-3D models often struggle with scenes containing multiple objects due to biases and occlusion complexities. To address this challenge, we present REPARO, a novel approach for compositional 3D asset generation from single images. REPARO employs a two-step process: first, it extracts individual objects from the scene and reconstructs their 3D meshes using off-the-shelf image-to-3D models; then, it optimizes the layout of these meshes through differentiable rendering techniques, ensuring coherent scene composition. By integrating optimal transport-based long-range appearance loss term and high-level semantic loss term in the differentiable rendering, REPARO can effectively recover the layout of 3D assets. The proposed method can significantly enhance object independence, detail accuracy, and overall scene coherence. Extensive evaluation of multi-object scenes demonstrates that our REPARO offers a comprehensive approach to address the complexities of multi-object 3D scene generation from single images.