Saved in:
Bibliographic Details
Main Authors: Xu, Chenyang, Deng, XingGuo, Zhong, Rui
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.22324
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908745390358528
author Xu, Chenyang
Deng, XingGuo
Zhong, Rui
author_facet Xu, Chenyang
Deng, XingGuo
Zhong, Rui
contents The 3D Gaussian Splatting (3D-GS) is a novel method for scene representation and view synthesis. Although Scaffold-GS achieves higher quality real-time rendering compared to the original 3D-GS, its fine-grained rendering of the scene is extremely dependent on adequate viewing angles. The spectral bias of neural network learning results in Scaffold-GS's poor ability to perceive and learn high-frequency information in the scene. In this work, we propose enhancing the manifold complexity of input features and using network-based feature map loss to improve the image reconstruction quality of 3D-GS models. We introduce AH-GS, which enables 3D Gaussians in structurally complex regions to obtain higher-frequency encodings, allowing the model to more effectively learn the high-frequency information of the scene. Additionally, we incorporate high-frequency reinforce loss to further enhance the model's ability to capture detailed frequency information. Our result demonstrates that our model significantly improves rendering fidelity, and in specific scenarios (e.g., MipNeRf360-garden), our method exceeds the rendering quality of Scaffold-GS in just 15K iterations.
format Preprint
id arxiv_https___arxiv_org_abs_2503_22324
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AH-GS: Augmented 3D Gaussian Splatting for High-Frequency Detail Representation
Xu, Chenyang
Deng, XingGuo
Zhong, Rui
Computer Vision and Pattern Recognition
Artificial Intelligence
The 3D Gaussian Splatting (3D-GS) is a novel method for scene representation and view synthesis. Although Scaffold-GS achieves higher quality real-time rendering compared to the original 3D-GS, its fine-grained rendering of the scene is extremely dependent on adequate viewing angles. The spectral bias of neural network learning results in Scaffold-GS's poor ability to perceive and learn high-frequency information in the scene. In this work, we propose enhancing the manifold complexity of input features and using network-based feature map loss to improve the image reconstruction quality of 3D-GS models. We introduce AH-GS, which enables 3D Gaussians in structurally complex regions to obtain higher-frequency encodings, allowing the model to more effectively learn the high-frequency information of the scene. Additionally, we incorporate high-frequency reinforce loss to further enhance the model's ability to capture detailed frequency information. Our result demonstrates that our model significantly improves rendering fidelity, and in specific scenarios (e.g., MipNeRf360-garden), our method exceeds the rendering quality of Scaffold-GS in just 15K iterations.
title AH-GS: Augmented 3D Gaussian Splatting for High-Frequency Detail Representation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2503.22324