PAOLI: Pose-free Articulated Object Learning from Sparse-view Images

2025-09-04

AI Summary

This paper addresses the problem of self-supervised learning of geometric and kinematic representations for articulated objects from only four sparse, unposed images per object. Methodologically, it introduces a progressive disentanglement framework that separately models static and dynamic parts while decoupling camera motion from joint articulation; it jointly optimizes sparse-view 3D reconstruction, deformable field modeling, and self-supervised consistency constraints across views and poses. To the best of our knowledge, this is the first approach to learn articulated object representations without pose supervision or dense multi-view inputs. Evaluated on standard benchmarks and real-world scenes, the method produces high-fidelity, fine-grained geometry–kinematics joint representations. It significantly reduces data dependency and outperforms prior methods in both reconstruction accuracy and kinematic plausibility.

Technology Category

• Computer Vision: Representation Learning for Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Unsupervised & Self-Supervised Learning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

We present a novel self-supervised framework for learning articulated object representations from sparse-view, unposed images. Unlike prior methods that require dense multi-view observations and ground-truth camera poses, our approach operates with as few as four views per articulation and no camera supervision. To address the inherent challenges, we first reconstruct each articulation independently using recent advances in sparse-view 3D reconstruction, then learn a deformation field that establishes dense correspondences across poses. A progressive disentanglement strategy further separates static from moving parts, enabling robust separation of camera and object motion. Finally, we jointly optimize geometry, appearance, and kinematics with a self-supervised loss that enforces cross-view and cross-pose consistency. Experiments on the standard benchmark and real-world examples demonstrate that our method produces accurate and detailed articulated object representations under significantly weaker input assumptions than existing approaches.

Problem

• Learn articulated object representations from sparse unposed images
• Reconstruct each articulation independently using sparse views
• Separate static and moving parts without camera supervision

Innovation

• Sparse-view 3D reconstruction for independent articulations
• Progressive disentanglement strategy separating static parts
• Self-supervised loss enforcing cross-view consistency