Articulat3D: Reconstructing Articulated Digital Twins From Monocular Videos with Geometric and Motion Constraints

TL;DR

Articulat3D introduces a novel framework for reconstructing high-fidelity digital twins of articulated objects from casual monocular videos by leveraging geometric and motion constraints, enabling scalable and accurate 3D modeling.

Contribution

The paper presents a new method that jointly enforces geometric and motion constraints for 3D reconstruction from monocular videos, including a motion prior-driven initialization and a refinement process with learnable kinematic primitives.

Findings

Achieves state-of-the-art results on synthetic benchmarks.

Successfully reconstructs articulated objects from casual monocular videos.

Demonstrates robustness in real-world uncontrolled conditions.

Abstract

Building high-fidelity digital twins of articulated objects from visual data remains a central challenge. Existing approaches depend on multi-view captures of the object in discrete, static states, which severely constrains their real-world scalability. In this paper, we introduce Articulat3D, a novel framework that constructs such digital twins from casually captured monocular videos by jointly enforcing explicit 3D geometric and motion constraints. We first propose Motion Prior-Driven Initialization, which leverages 3D point tracks to exploit the low-dimensional structure of articulated motion. By modeling scene dynamics with a compact set of motion bases, we facilitate soft decomposition of the scene into multiple rigidly-moving groups. Building on this initialization, we introduce Geometric and Motion Constraints Refinement, which enforces physically plausible articulation through learnable kinematic primitives parameterized by a joint axis, a pivot point, and per-frame motion scalars, yielding reconstructions that are both geometrically accurate and temporally coherent. Extensive experiments demonstrate that Articulat3D achieves state-of-the-art performance on synthetic benchmarks and real-world casually captured monocular videos, significantly advancing the feasibility of digital twin creation under uncontrolled real-world conditions. Our project page is at https://maxwell-zhao.github.io/Articulat3D.