GeoRect4D: Geometry-Compatible Generative Rectification for Dynamic Sparse-View 3D Reconstruction

2026-04-22

AI Summary

This work addresses the ill-posed challenges in dynamic 3D scene reconstruction from sparse viewpoints—such as geometric collapse, trajectory drift, and floating artifacts—by introducing the GeoRect4D framework. The proposed method integrates explicit 3D consistency constraints with generative refinement, leveraging a structure-locking mechanism and spatiotemporal-coordinated attention to ensure geometric compatibility of synthesized content. Degradation-aware feedback and a progressive optimization strategy jointly eliminate floating artifacts while enhancing texture details. Further improvements are achieved through an anchor-based dynamic 3D Gaussian Splatting representation, a single-step diffusion corrector, and generative distillation. Experiments demonstrate that GeoRect4D significantly outperforms existing approaches across multiple datasets, achieving state-of-the-art performance in reconstruction fidelity, perceptual quality, and spatiotemporal consistency.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

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.

Problem

• dynamic 3D reconstruction
• sparse-view
• geometric consistency
• temporal inconsistency
• generative priors

Innovation

• dynamic 3D reconstruction
• generative rectification
• geometry-consistent generation
• spatiotemporal attention
• 3D Gaussian Splatting