LGDWT-GS: Local and Global Discrete Wavelet-Regularized 3D Gaussian Splatting for Sparse-View Scene Reconstruction

2026-01-23

AI Summary

This work addresses geometric instability and high-frequency detail loss in 3D Gaussian splatting under sparse-view settings by introducing multi-scale discrete wavelet regularization into the framework for the first time. By jointly enforcing local and global wavelet constraints, the proposed method enhances geometric consistency while preserving fine-scale details. The core contributions include the first 3D Gaussian splatting approach integrated with multi-scale wavelet regularization and the creation of the first greenhouse plant dataset spanning four spectral bands, accompanied by a few-shot reconstruction benchmark. Experimental results demonstrate that the method achieves sharper, more stable, and spectrally consistent reconstructions on both the newly introduced multispectral dataset and standard benchmarks, significantly outperforming existing approaches.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

We propose a new method for few-shot 3D reconstruction that integrates global and local frequency regularization to stabilize geometry and preserve fine details under sparse-view conditions, addressing a key limitation of existing 3D Gaussian Splatting (3DGS) models. We also introduce a new multispectral greenhouse dataset containing four spectral bands captured from diverse plant species under controlled conditions. Alongside the dataset, we release an open-source benchmarking package that defines standardized few-shot reconstruction protocols for evaluating 3DGS-based methods. Experiments on our multispectral dataset, as well as standard benchmarks, demonstrate that the proposed method achieves sharper, more stable, and spectrally consistent reconstructions than existing baselines. The dataset and code for this work are publicly available

Problem

• 3D Gaussian Splatting
• sparse-view reconstruction
• geometry stability
• fine detail preservation
• few-shot 3D reconstruction

Innovation

• Discrete Wavelet Regularization
• 3D Gaussian Splatting
• Sparse-View Reconstruction
• Multispectral Dataset
• Few-Shot 3D Reconstruction