让文字引导3D模型补全被遮挡部分，保持真实观测不变
RelaxFlow: Text-Driven Amodal 3D Generation
- 分两路控制：观察部分严格约束，文字提示区域灵活生成
- 在极端遮挡下仍能准确匹配文本意图，保持视觉质量
- 无需训练，适合快速生成复杂场景的3D内容
图像到3D生成在遮挡情况下存在固有的语义模糊性，仅靠部分观测往往不足以确定物体类别。本文提出文本驱动的非可视3D生成任务，即在严格保留输入观测的前提下，由文本提示引导未见区域的补全。关键发现是：该任务需要不同粒度的控制机制——对观测部分需刚性控制，对提示部分则需松弛的结构控制。为此，我们提出 RelaxFlow，一个无需训练的双分支框架，通过多先验共识模块与松弛机制解耦控制粒度。理论上证明，该松弛等价于对生成向量场施加低通滤波，抑制高频实例细节，保留可容纳观测的几何结构。为促进评估，我们引入两个诊断基准：ExtremeOcc-3D 和 AmbiSem-3D。大量实验表明，RelaxFlow 能有效引导未见区域生成以匹配文本意图，且不损害视觉保真度。
原文摘要 · Abstract (English)
Image-to-3D generation faces inherent semantic ambiguity under occlusion, where partial observation alone is often insufficient to determine object category. In this work, we formalize text-driven amodal 3D generation, where text prompts steer the completion of unseen regions while strictly preserving input observation. Crucially, we identify that these objectives demand distinct control granularities: rigid control for the observation versus relaxed structural control for the prompt. To this end, we propose RelaxFlow, a training-free dual-branch framework that decouples control granularity via a Multi-Prior Consensus Module and a Relaxation Mechanism. Theoretically, we prove that our relaxation is equivalent to applying a low-pass filter on the generative vector field, which suppresses high-frequency instance details to isolate geometric structure that accommodates the observation. To facilitate evaluation, we introduce two diagnostic benchmarks, ExtremeOcc-3D and AmbiSem-3D. Extensive experiments demonstrate that RelaxFlow successfully steers the generation of unseen regions to match the prompt intent without compromising visual fidelity.
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。