Frequency-Modulated Point Cloud Rendering with Easy Editing

TL;DR

This paper introduces a novel point cloud rendering pipeline that combines adaptive frequency modulation with geometry optimization, enabling high-fidelity, real-time view synthesis and easy editing, outperforming existing methods.

Contribution

The paper presents AFNet, an adaptive frequency modulation module using a hypernetwork, and a point cloud preprocessing step, enhancing rendering quality and editing capabilities with low computational cost.

Findings

Achieves superior PSNR, SSIM, LPIPS on multiple datasets.

Supports real-time rendering and interactive editing.

Outperforms state-of-the-art methods in quality metrics.

Abstract

We develop an effective point cloud rendering pipeline for novel view synthesis, which enables high fidelity local detail reconstruction, real-time rendering and user-friendly editing. In the heart of our pipeline is an adaptive frequency modulation module called Adaptive Frequency Net (AFNet), which utilizes a hypernetwork to learn the local texture frequency encoding that is consecutively injected into adaptive frequency activation layers to modulate the implicit radiance signal. This mechanism improves the frequency expressive ability of the network with richer frequency basis support, only at a small computational budget. To further boost performance, a preprocessing module is also proposed for point cloud geometry optimization via point opacity estimation. In contrast to implicit rendering, our pipeline supports high-fidelity interactive editing based on point cloud manipulation. Extensive experimental results on NeRF-Synthetic, ScanNet, DTU and Tanks and Temples datasets demonstrate the superior performances achieved by our method in terms of PSNR, SSIM and LPIPS, in comparison to the state-of-the-art.