We present PAD3R, a method for reconstructing deformable 3D objects from casually captured, unposed monocular videos. Unlike existing approaches, PAD3R handles long video sequences featuring substantial object deformation, large-scale camera movement, and limited view coverage that typically challenge conventional systems. At its core, our approach trains a personalized, object-centric pose estimator, supervised by a pre-trained image-to-3D model. This guides the optimization of deformable 3D Gaussian representation. The optimization is further regularized by long-term 2D point tracking over the entire input video. By combining generative priors and differentiable rendering, PAD3R reconstructs high-fidelity, articulated 3D representations of objects in a category-agnostic way. Extensive qualitative and quantitative results show that PAD3R is robust and generalizes well across challenging scenarios, highlighting its potential for dynamic scene understanding and 3D content creation.
@article{arxiv.2509.25183,
title = {PAD3R: Pose-Aware Dynamic 3D Reconstruction from Casual Videos},
author = {Ting-Hsuan Liao and Haowen Liu and Yiran Xu and Songwei Ge and Gengshan Yang and Jia-Bin Huang},
journal= {arXiv preprint arXiv:2509.25183},
year = {2025}
}
Comments
SIGGRAPH Asia 2025. Project page:https://pad3r.github.io/