Orientation-anchored Hyper-Gaussian for 4D Reconstruction from Casual Videos

2025-09-27

AI Summary

Existing monocular video 4D reconstruction methods rely on low-rank assumptions, limiting their ability to model complex, spatially heterogeneous dynamic deformations. To address this, we propose Oriented Hyper-Gaussian Representation (OHGR), the first method to explicitly embed a global scene orientation field as a core latent state in 4D modeling—thereby eliminating restrictive low-rank constraints. OHGR jointly models spatiotemporal, geometric, and orientational information through orientation-guided geometric evolution and conditional slicing, enabling region-adaptive deformation inference. Evaluated on real-world complex dynamic scenes, our approach achieves significantly higher reconstruction fidelity compared to state-of-the-art methods, demonstrating superior accuracy in geometry, motion, and structural coherence.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Computer Vision: 3D Computer Vision
• Machine Learning: Representation Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

We present Orientation-anchored Gaussian Splatting (OriGS), a novel framework for high-quality 4D reconstruction from casually captured monocular videos. While recent advances extend 3D Gaussian Splatting to dynamic scenes via various motion anchors, such as graph nodes or spline control points, they often rely on low-rank assumptions and fall short in modeling complex, region-specific deformations inherent to unconstrained dynamics. OriGS addresses this by introducing a hyperdimensional representation grounded in scene orientation. We first estimate a Global Orientation Field that propagates principal forward directions across space and time, serving as stable structural guidance for dynamic modeling. Built upon this, we propose Orientation-aware Hyper-Gaussian, a unified formulation that embeds time, space, geometry, and orientation into a coherent probabilistic state. This enables inferring region-specific deformation through principled conditioned slicing, adaptively capturing diverse local dynamics in alignment with global motion intent. Experiments demonstrate the superior reconstruction fidelity of OriGS over mainstream methods in challenging real-world dynamic scenes.

Problem

• Modeling complex deformations in dynamic 4D reconstruction
• Capturing region-specific motion from unconstrained monocular videos
• Inferring local dynamics aligned with global structural guidance

Innovation

• Orientation-anchored Gaussian Splatting for 4D reconstruction
• Global Orientation Field guides dynamic scene modeling
• Orientation-aware Hyper-Gaussian embeds unified probabilistic state