RelaxFlow: Text-Driven Amodal 3D Generation
This work addresses the semantic ambiguity under occlusion in image-to-3D generation, which is a problem for anyone trying to create complete 3D models from partial observations.
This paper tackles the problem of text-driven amodal 3D generation, where text prompts guide the completion of occluded regions while strictly preserving observed parts. They propose RelaxFlow, a training-free dual-branch framework that successfully steers the generation of unseen regions to match the prompt intent without compromising visual fidelity.
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.