Neural Gabor Splatting: Enhanced Gaussian Splatting with Neural Gabor for High-frequency Surface Reconstruction

2026-04-17

AI Summary

This work addresses the limitation of 3D Gaussian splatting in reconstruct日晚间高频 appearance details, where uniform per-primitive colors necessitate an excessive number of primitives to capture sharp color transitions. To overcome this, the authors propose Neural Gabor Splatting, which equips each Gaussian primitive with a lightweight MLP to model intricate internal color variations and incorporates Gabor features to enhance frequency representation. Additionally, a frequency-aware primitive densification and pruning mechanism is introduced to dynamically regulate scene density. Evaluated on Mip-NeRF360 and high-frequency datasets such as checkerboard patterns, the method significantly outperforms existing baselines, achieving superior reconstruction fidelity with substantially fewer primitives—demonstrating a synergistic improvement in both accuracy and efficiency.

Technology Category

• Computer Vision: 3D Computer Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Recent years have witnessed the rapid emergence of 3D Gaussian splatting (3DGS) as a powerful approach for 3D reconstruction and novel view synthesis. Its explicit representation with Gaussian primitives enables fast training, real-time rendering, and convenient post-processing such as editing and surface reconstruction. However, 3DGS suffers from a critical drawback: the number of primitives grows drastically for scenes with high-frequency appearance details, since each primitive can represent only a single color, requiring multiple primitives for every sharp color transition. To overcome this limitation, we propose neural Gabor splatting, which augments each Gaussian primitive with a lightweight multi-layer perceptron that models a wide range of color variations within a single primitive. To further control primitive numbers, we introduce a frequency-aware densification strategy that selects mismatch primitives for pruning and cloning based on frequency energy. Our method achieves accurate reconstruction of challenging high-frequency surfaces. We demonstrate its effectiveness through extensive experiments on both standard benchmarks, such as Mip-NeRF360 and High-Frequency datasets (e.g., checkered patterns), supported by comprehensive ablation studies.

Problem

• 3D Gaussian splatting
• high-frequency surface reconstruction
• color transition
• primitive representation

Innovation

• Neural Gabor Splatting
• 3D Gaussian Splatting
• high-frequency reconstruction
• frequency-aware densification
• learned color variation