Geometry-aware Reconstruction and Fusion-refined Rendering for Generalizable Neural Radiance Fields

TL;DR

This paper introduces GeFu, a novel framework for generalizable neural radiance fields that enhances geometry accuracy and view synthesis quality through adaptive cost aggregation, spatial-view descriptor aggregation, and a fusion strategy.

Contribution

The paper proposes ACA, SVA, and CAF modules integrated into a coarse-to-fine framework to improve generalization and rendering quality of NeRFs in challenging scenarios.

Findings

Achieves state-of-the-art results on multiple datasets.

Improves geometry consistency across views.

Enhances novel view synthesis quality.

Abstract

Generalizable NeRF aims to synthesize novel views for unseen scenes. Common practices involve constructing variance-based cost volumes for geometry reconstruction and encoding 3D descriptors for decoding novel views. However, existing methods show limited generalization ability in challenging conditions due to inaccurate geometry, sub-optimal descriptors, and decoding strategies. We address these issues point by point. First, we find the variance-based cost volume exhibits failure patterns as the features of pixels corresponding to the same point can be inconsistent across different views due to occlusions or reflections. We introduce an Adaptive Cost Aggregation (ACA) approach to amplify the contribution of consistent pixel pairs and suppress inconsistent ones. Unlike previous methods that solely fuse 2D features into descriptors, our approach introduces a Spatial-View Aggregator (SVA) to incorporate 3D context into descriptors through spatial and inter-view interaction. When decoding the descriptors, we observe the two existing decoding strategies excel in different areas, which are complementary. A Consistency-Aware Fusion (CAF) strategy is proposed to leverage the advantages of both. We incorporate the above ACA, SVA, and CAF into a coarse-to-fine framework, termed Geometry-aware Reconstruction and Fusion-refined Rendering (GeFu). GeFu attains state-of-the-art performance across multiple datasets. Code is available at https://github.com/TQTQliu/GeFu .