ConMax3D: Frame Selection for 3D Reconstruction Through Concept Maximization

Akash Malhotra, Akash Malhotra, Nacéra Seghouani, Gilbert Badaro, Christophe Blaya

2025

Abstract

This paper proposes a novel best frames selection algorithm, ConMax3D, for multiview 3D reconstruction that utilizes image segmentation and clustering to identify and maximize concept diversity. This method aims to improve the accuracy and interpretability of selecting frames for a photorealistic 3D model generation with NeRF or 3D Gaussian Splatting without relying on camera pose information. We evaluate ConMax3D on the LLFF dataset and show that it outperforms current state-of-the-art baselines, with improvements in PSNR of up to 43.65%, while retaining computational efficiency.

Paper Citation

in Harvard Style

Malhotra A., Seghouani N., Badaro G. and Blaya C. (2025). ConMax3D: Frame Selection for 3D Reconstruction Through Concept Maximization. In Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP; ISBN 978-989-758-728-3, SciTePress, pages 598-609. DOI: 10.5220/0013258800003912

in Bibtex Style

@conference{visapp25,
author={Akash Malhotra and Nacéra Seghouani and Gilbert Badaro and Christophe Blaya},
title={ConMax3D: Frame Selection for 3D Reconstruction Through Concept Maximization},
booktitle={Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP},
year={2025},
pages={598-609},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013258800003912},
isbn={978-989-758-728-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP
TI - ConMax3D: Frame Selection for 3D Reconstruction Through Concept Maximization
SN - 978-989-758-728-3
AU - Malhotra A.
AU - Seghouani N.
AU - Badaro G.
AU - Blaya C.
PY - 2025
SP - 598
EP - 609
DO - 10.5220/0013258800003912
PB - SciTePress