Lie Flow: Video Dynamic Fields Modeling and Predicting with Lie Algebra as Geometric Physics Principle

TL;DR

LieFlow introduces a novel 4D scene modeling framework that explicitly incorporates Lie algebra within the SE(3) group to better capture complex motions, improving physical realism and view synthesis in dynamic scenes.

Contribution

The paper proposes LieFlow, a framework that models 4D scenes using SE(3) Lie group transformations, unifying translation and rotation for physically consistent dynamic scene representation.

Findings

Consistently improves view-synthesis quality over baselines.

Enhances temporal coherence and physical realism in dynamic scene modeling.

Demonstrates robustness on synthetic and real-world datasets.

Abstract

Modeling 4D scenes requires capturing both spatial structure and temporal motion, which is challenging due to the need for physically consistent representations of complex rigid and non-rigid motions. Existing approaches mainly rely on translational displacements, which struggle to represent rotations, articulated transformations, often leading to spatial inconsistency and physically implausible motion. LieFlow, a dynamic radiance representation framework that explicitly models motion within the SE(3) Lie group, enabling coherent learning of translation and rotation in a unified geometric space. The SE(3) transformation field enforces physically inspired constraints to maintain motion continuity and geometric consistency. The evaluation includes a synthetic dataset with rigid-body trajectories and two real-world datasets capturing complex motion under natural lighting and occlusions. Across all datasets, LieFlow consistently improves view-synthesis fidelity, temporal coherence, and physical realism over NeRF-based baselines. These results confirm that SE(3)-based motion modeling offers a robust and physically grounded framework for representing dynamic 4D scenes.