Saved in:
Bibliographic Details
Main Authors: Xu, Bo, Guo, Yuhu, Wang, Yuchao, Wang, Wenting, Yam, Yeung, Wang, Charlie C. L., Le, Xinyi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.18314
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915462846087168
author Xu, Bo
Guo, Yuhu
Wang, Yuchao
Wang, Wenting
Yam, Yeung
Wang, Charlie C. L.
Le, Xinyi
author_facet Xu, Bo
Guo, Yuhu
Wang, Yuchao
Wang, Wenting
Yam, Yeung
Wang, Charlie C. L.
Le, Xinyi
contents We propose a semantic-aware neural reconstruction method to generate 3D high-fidelity models from sparse images. To tackle the challenge of severe radiance ambiguity caused by mismatched features in sparse input, we enrich neural implicit representations by adding patch-based semantic logits that are optimized together with the signed distance field and the radiance field. A novel regularization based on the geometric primitive masks is introduced to mitigate shape ambiguity. The performance of our approach has been verified in experimental evaluation. The average chamfer distances of our reconstruction on the DTU dataset can be reduced by 44% for SparseNeuS and 20% for VolRecon. When working as a plugin for those dense reconstruction baselines such as NeuS and Neuralangelo, the average error on the DTU dataset can be reduced by 69% and 68% respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2508_18314
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SERES: Semantic-aware neural reconstruction from sparse views
Xu, Bo
Guo, Yuhu
Wang, Yuchao
Wang, Wenting
Yam, Yeung
Wang, Charlie C. L.
Le, Xinyi
Computer Vision and Pattern Recognition
We propose a semantic-aware neural reconstruction method to generate 3D high-fidelity models from sparse images. To tackle the challenge of severe radiance ambiguity caused by mismatched features in sparse input, we enrich neural implicit representations by adding patch-based semantic logits that are optimized together with the signed distance field and the radiance field. A novel regularization based on the geometric primitive masks is introduced to mitigate shape ambiguity. The performance of our approach has been verified in experimental evaluation. The average chamfer distances of our reconstruction on the DTU dataset can be reduced by 44% for SparseNeuS and 20% for VolRecon. When working as a plugin for those dense reconstruction baselines such as NeuS and Neuralangelo, the average error on the DTU dataset can be reduced by 69% and 68% respectively.
title SERES: Semantic-aware neural reconstruction from sparse views
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.18314