Drantal-NeRF: Diffusion-Based Restoration for Anti-aliasing Neural Radiance Field

TL;DR

This paper introduces Drantal-NeRF, a diffusion-based method that effectively reduces aliasing artifacts in NeRF renderings by treating aliasing as a degradation and restoring high-quality images, improving realism and consistency.

Contribution

The paper proposes a novel diffusion-based restoration approach for anti-aliasing in NeRF, incorporating multi-view consistency and scene-specific finetuning, which is simple, versatile, and improves rendering quality.

Findings

Significant qualitative improvements in large-scale urban scenes.

Quantitative metrics show enhanced anti-aliasing performance.

Method is compatible with various NeRF architectures.

Abstract

Aliasing artifacts in renderings produced by Neural Radiance Field (NeRF) is a long-standing but complex issue in the field of 3D implicit representation, which arises from a multitude of intricate causes and was mitigated by designing more advanced but complex scene parameterization methods before. In this paper, we present a Diffusion-based restoration method for anti-aliasing Neural Radiance Field (Drantal-NeRF). We consider the anti-aliasing issue from a low-level restoration perspective by viewing aliasing artifacts as a kind of degradation model added to clean ground truths. By leveraging the powerful prior knowledge encapsulated in diffusion model, we could restore the high-realism anti-aliasing renderings conditioned on aliased low-quality counterparts. We further employ a feature-wrapping operation to ensure multi-view restoration consistency and finetune the VAE decoder to better adapt to the scene-specific data distribution. Our proposed method is easy to implement and agnostic to various NeRF backbones. We conduct extensive experiments on challenging large-scale urban scenes as well as unbounded 360-degree scenes and achieve substantial qualitative and quantitative improvements.