Saved in:
Bibliographic Details
Main Authors: Sui, Jialu, Liu, Rui, Zhang, Hongsheng
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.08506
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917135175909376
author Sui, Jialu
Liu, Rui
Zhang, Hongsheng
author_facet Sui, Jialu
Liu, Rui
Zhang, Hongsheng
contents A major challenge in reconstructing buildings from LiDAR point clouds lies in accurately capturing building surfaces under varying point densities and noise interference. To flexibly gather high-quality 3D profiles of the building in diverse resolution, we propose OCCDiff applying latent diffusion in the occupancy function space. Our OCCDiff combines a latent diffusion process with a function autoencoder architecture to generate continuous occupancy functions evaluable at arbitrary locations. Moreover, a point encoder is proposed to provide condition features to diffusion learning, constraint the final occupancy prediction for occupancy decoder, and insert multi-modal features for latent generation to latent encoder. To further enhance the model performance, a multi-task training strategy is employed, ensuring that the point encoder learns diverse and robust feature representations. Empirical results show that our method generates physically consistent samples with high fidelity to the target distribution and exhibits robustness to noisy data.
format Preprint
id arxiv_https___arxiv_org_abs_2512_08506
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OCCDiff: Occupancy Diffusion Model for High-Fidelity 3D Building Reconstruction from Noisy Point Clouds
Sui, Jialu
Liu, Rui
Zhang, Hongsheng
Computer Vision and Pattern Recognition
A major challenge in reconstructing buildings from LiDAR point clouds lies in accurately capturing building surfaces under varying point densities and noise interference. To flexibly gather high-quality 3D profiles of the building in diverse resolution, we propose OCCDiff applying latent diffusion in the occupancy function space. Our OCCDiff combines a latent diffusion process with a function autoencoder architecture to generate continuous occupancy functions evaluable at arbitrary locations. Moreover, a point encoder is proposed to provide condition features to diffusion learning, constraint the final occupancy prediction for occupancy decoder, and insert multi-modal features for latent generation to latent encoder. To further enhance the model performance, a multi-task training strategy is employed, ensuring that the point encoder learns diverse and robust feature representations. Empirical results show that our method generates physically consistent samples with high fidelity to the target distribution and exhibits robustness to noisy data.
title OCCDiff: Occupancy Diffusion Model for High-Fidelity 3D Building Reconstruction from Noisy Point Clouds
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.08506