HI²: Sparse-View 3D Object Reconstruction with a Hybrid Implicit Initialization

Pragati Jaiswal, Pragati Jaiswal, Didier Stricker, Didier Stricker

2025

Abstract

Accurate 3D object reconstruction is essential for various applications, including mixed reality and medicine. Recent advancements in deep learning-based methods and implicit 3D modelling have significantly enhanced the accuracy of 3D object reconstruction. Traditional methods enable reconstruction from a limited number of images, while implicit 3D modelling is proficient at capturing fine details and complex topologies. In this paper, we present a novel pipeline for 3D object reconstruction that combines the strengths of both approaches. Firstly, we use a 3D occupancy grid to generate a coarse 3D object from a few images. Secondly, we implement a novel and effective sampling strategy to transform the coarse reconstruction into an implicit representation, which is optimized to reduce computation power and training time. This sampling strategy also allows it to be true to scale given actual camera intrinsic and extrinsic parameters. Finally, we refine the implicit representation and extract the 3D object mesh under a differentiable rendering scheme. Experiments on several datasets demonstrate that our proposed approach can reconstruct accurate 3D objects and outperforms state-of-the-art methods in terms of the Chamfer distance and Peak Signal-to-Noise Ratio metrics.

Paper Citation

in Harvard Style

Jaiswal P. and Stricker D. (2025). HI²: Sparse-View 3D Object Reconstruction with a Hybrid Implicit Initialization. In Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-730-6, SciTePress, pages 170-180. DOI: 10.5220/0013382000003905

in Bibtex Style

@conference{icpram25,
author={Pragati Jaiswal and Didier Stricker},
title={HI²: Sparse-View 3D Object Reconstruction with a Hybrid Implicit Initialization},
booktitle={Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2025},
pages={170-180},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013382000003905},
isbn={978-989-758-730-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - HI²: Sparse-View 3D Object Reconstruction with a Hybrid Implicit Initialization
SN - 978-989-758-730-6
AU - Jaiswal P.
AU - Stricker D.
PY - 2025
SP - 170
EP - 180
DO - 10.5220/0013382000003905
PB - SciTePress