SceneMotifCoder: Example-driven Visual Program Learning for Generating 3D Object Arrangements

TL;DR

SceneMotifCoder (SMC) is a novel example-driven framework that uses visual program learning, large language models, and geometry-aware optimization to generate diverse, plausible 3D object arrangements aligned with text descriptions.

Contribution

Introduces SceneMotifCoder, a new method combining program synthesis, large language models, and 3D retrieval to improve multi-object arrangement generation from few examples.

Findings

SMC produces high-quality, diverse 3D arrangements.

Arrangements better match user text descriptions.

Generated arrangements are more physically plausible.

Abstract

Despite advances in text-to-3D generation methods, generation of multi-object arrangements remains challenging. Current methods exhibit failures in generating physically plausible arrangements that respect the provided text description. We present SceneMotifCoder (SMC), an example-driven framework for generating 3D object arrangements through visual program learning. SMC leverages large language models (LLMs) and program synthesis to overcome these challenges by learning visual programs from example arrangements. These programs are generalized into compact, editable meta-programs. When combined with 3D object retrieval and geometry-aware optimization, they can be used to create object arrangements varying in arrangement structure and contained objects. Our experiments show that SMC generates high-quality arrangements using meta-programs learned from few examples. Evaluation results demonstrates that object arrangements generated by SMC better conform to user-specified text descriptions and are more physically plausible when compared with state-of-the-art text-to-3D generation and layout methods.