RelaxFlow: Text-Driven Amodal 3D Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Jiayin, Fu, Guoji, Liu, Xiaolu, He, Qiyuan, Li, Yicong, Yao, Angela
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917539620061184
author Zhu, Jiayin
Fu, Guoji
Liu, Xiaolu
He, Qiyuan
Li, Yicong
Yao, Angela
author_facet Zhu, Jiayin
Fu, Guoji
Liu, Xiaolu
He, Qiyuan
Li, Yicong
Yao, Angela
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_05425
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RelaxFlow: Text-Driven Amodal 3D Generation
Zhu, Jiayin
Fu, Guoji
Liu, Xiaolu
He, Qiyuan
Li, Yicong
Yao, Angela
Computer Vision and Pattern Recognition
Artificial Intelligence
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.
title RelaxFlow: Text-Driven Amodal 3D Generation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2603.05425