Search papers, labs, and topics across Lattice.
This paper introduces Flux-GS, a novel real-time Gaussian Splatting method that significantly reduces the inference and storage overhead associated with high-order Spherical Harmonics for mobile platforms. By employing a Monte Carlo Specular Energy Aggregator and an Attribute-Conditioned SH Enhancement module, Flux-GS effectively preserves critical lighting features while minimizing high-frequency detail loss during compression. Experimental results reveal that Flux-GS achieves a substantial reduction in parameters while maintaining competitive visual quality, making it a viable solution for mobile rendering applications.
Flux-GS slashes the storage overhead for mobile 3D rendering while preserving essential visual fidelity, enabling high-quality graphics on resource-constrained devices.
Recent advances in 3D Gaussian Splatting have demonstrated unprecedented success in novel view synthesis. However, the substantial inference and storage overhead driven by high-order Spherical Harmonics (SH) are primary bottlenecks for mobile platforms. In this paper, we present Flux-GS, a real-time Gaussian Splatting method designed to achieve high-fidelity rendering with significantly reduced overhead for resource-constrained mobile platforms. We first propose a Monte Carlo Specular Energy Aggregator, sampling third-order radiance residuals and aggregating specular energy into a compact latent space. In this way, our method effectively preserves visually salient lighting features in lower-order bands without expensive distillation or pre-training. To mitigate the high-frequency details lost during compression, we introduce an Attribute-Conditioned SH Enhancement module. This module predicts Gaussian-aware offsets based on intrinsic Gaussian attributes, which enhance the first-order SH representation prior to inference, without extra inference costs. Furthermore, the original single-view gradient-based densification is prone to producing excessive Gaussians and overfitting to a certain view. We address these limitations by proposing a Multi-view Alpha-based Densification and Pruning strategy. By leveraging multi-view guidance, we ensure multi-view structure consistency and the precise removal of redundant primitives. Extensive experiments demonstrate that Flux-GS achieves substantial parameter reduction while maintaining competitive visual quality, offering a robust and scalable solution for real-time mobile rendering. Code: magenta{https://xiaobiaodu.github.io/flux-gs-project/{https://xiaobiaodu.github.io/flux-gs-project/}}.
