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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Deng, Jianning, Subr, Kartic, Bilen, Hakan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914441219538944
author Deng, Jianning
Subr, Kartic
Bilen, Hakan
author_facet Deng, Jianning
Subr, Kartic
Bilen, Hakan
contents We present a methodology to model articulated objects using a sparse set of images with unknown poses. Current methods 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. Our central insight is to first solve a robust correspondence and alignment problem between unaligned reconstructions, before part motions can be analyzed. 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 optimize geometry, appearance, and kinematics jointly 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.
format Preprint
id arxiv_https___arxiv_org_abs_2509_04276
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PAOLI: Pose-free Articulated Object Learning from Sparse-view Images
Deng, Jianning
Subr, Kartic
Bilen, Hakan
Computer Vision and Pattern Recognition
We present a methodology to model articulated objects using a sparse set of images with unknown poses. Current methods 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. Our central insight is to first solve a robust correspondence and alignment problem between unaligned reconstructions, before part motions can be analyzed. 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 optimize geometry, appearance, and kinematics jointly 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.
title PAOLI: Pose-free Articulated Object Learning from Sparse-view Images
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.04276