ReFlow: Self-correction Motion Learning for Dynamic Scene Reconstruction

TL;DR

ReFlow introduces a self-correcting framework for monocular dynamic scene reconstruction, improving initialization, decoupling static and dynamic components, and enforcing multi-view consistency for more accurate 4D reconstructions.

Contribution

The paper proposes a novel self-correction flow matching mechanism and modules for enhanced initialization and dynamic scene modeling, advancing monocular 4D reconstruction techniques.

Findings

ReFlow achieves superior reconstruction quality across diverse scenarios.

The self-correction paradigm improves robustness and accuracy in dynamic scene reconstruction.

Decoupling static and dynamic components enhances motion estimation stability.

Abstract

We present ReFlow, a unified framework for monocular dynamic scene reconstruction that learns 3D motion in a novel self-correction manner from raw video. Existing methods often suffer from incomplete scene initialization for dynamic regions, leading to unstable reconstruction and motion estimation, which often resorts to external dense motion guidance such as pre-computed optical flow to further stabilize and constrain the reconstruction of dynamic components. However, this introduces additional complexity and potential error propagation. To address these issues, ReFlow integrates a Complete Canonical Space Construction module for enhanced initialization of both static and dynamic regions, and a Separation-Based Dynamic Scene Modeling module that decouples static and dynamic components for targeted motion supervision. The core of ReFlow is a novel self-correction flow matching mechanism, consisting of Full Flow Matching to align 3D scene flow with time-varying 2D observations, and Camera Flow Matching to enforce multi-view consistency for static objects. Together, these modules enable robust and accurate dynamic scene reconstruction. Extensive experiments across diverse scenarios demonstrate that ReFlow achieves superior reconstruction quality and robustness, establishing a novel self-correction paradigm for monocular 4D reconstruction.