通过分阶段优化解决3D组合生成中的梯度冲突问题
DecompDreamer: A Composition-Aware Curriculum for Structured 3D Asset Generation
- 先建结构骨架，再细化单个组件，避免梯度冲突
- 在多个组合提示上实现更高保真度与空间一致性
- 适合需要精确布局的3D资产生成场景
当前文本到3D的方法在生成单一物体时表现优异，但在组合提示下表现不佳。我们认为这一失败源于其优化调度机制，因为同时或迭代的启发式方法在面对组合爆炸带来的冲突梯度时会不可避免地导致几何纠缠或灾难性发散。本文将组合生成的核心挑战重新定义为优化调度问题。我们提出 DecompDreamer，一种基于新型分阶段优化策略的框架，作为隐式课程。该方法首先通过优先处理对象间关系建立一致的结构骨架，随后转向单个组件的高保真度精细化。这种时间上的目标解耦提供了对梯度冲突的稳健解决方案。在多样化组合提示上的定性和定量评估表明，DecompDreamer 在保真度、解缠和空间连贯性方面均优于现有最先进方法。
原文摘要 · Abstract (English)
Current text-to-3D methods excel at generating single objects but falter on compositional prompts. We argue this failure is fundamental to their optimization schedules, as simultaneous or iterative heuristics predictably collapse under a combinatorial explosion of conflicting gradients, leading to entangled geometry or catastrophic divergence. In this paper, we reframe the core challenge of compositional generation as one of optimization scheduling. We introduce DecompDreamer, a framework built on a novel staged optimization strategy that functions as an implicit curriculum. Our method first establishes a coherent structural scaffold by prioritizing inter-object relationships before shifting to the high-fidelity refinement of individual components. This temporal decoupling of competing objectives provides a robust solution to gradient conflict. Qualitative and quantitative evaluations on diverse compositional prompts demonstrate that DecompDreamer outperforms state-of-the-art methods in fidelity, disentanglement, and spatial coherence.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。