GeoRect4D: Geometry-Compatible Generative Rectification for Dynamic Sparse-View 3D Reconstruction
2026-04-22 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors present GeoRect4D, a new method to create detailed 3D videos from a small number of camera views. They combine a reliable 3D base model with a smart correction step that adds missing details without messing up the scene's structure or timing. Their approach also includes ways to clean up errors and improve textures over time. Tests show that GeoRect4D makes more accurate and realistic 3D reconstructions compared to previous methods.
3D reconstructionmulti-view videosdynamic scenesgenerative priorsdiffusion modelsspatiotemporal consistencygeometric purificationtexture distillation3D geometryoptimization
Authors
Zhenlong Wu, Zihan Zheng, Xuanxuan Wang, Qianhe Wang, Hua Yang, Xiaoyun Zhang, Qiang Hu, Wenjun Zhang
Abstract
Reconstructing dynamic 3D scenes from sparse multi-view videos is highly ill-posed, often leading to geometric collapse, trajectory drift, and floating artifacts. Recent attempts introduce generative priors to hallucinate missing content, yet naive integration frequently causes structural drift and temporal inconsistency due to the mismatch between stochastic 2D generation and deterministic 3D geometry. In this paper, we propose GeoRect4D, a novel unified framework for sparse-view dynamic reconstruction that couples explicit 3D consistency with generative refinement via a closed-loop optimization process. Specifically, GeoRect4D introduces a degradation-aware feedback mechanism that incorporates a robust anchor-based dynamic 3DGS substrate with a single-step diffusion rectifier to hallucinate high-fidelity details. This rectifier utilizes a structural locking mechanism and spatiotemporal coordinated attention, effectively preserving physical plausibility while restoring missing content. Furthermore, we present a progressive optimization strategy that employs stochastic geometric purification to eliminate floaters and generative distillation to infuse texture details into the explicit representation. Extensive experiments demonstrate that GeoRect4D achieves state-of-the-art performance in reconstruction fidelity, perceptual quality, and spatiotemporal consistency across multiple datasets.