Learning Neural Duplex Radiance Fields for Real-Time View Synthesis

• Main Authors: Wan, Ziyu, Richardt, Christian, Božič, Aljaž, Li, Chao, Rengarajan, Vijay, Nam, Seonghyeon, Xiang, Xiaoyu, Li, Tuotuo, Zhu, Bo, Ranjan, Rakesh, Liao, Jing
• Format: Journal Article
• Language: English
• Published: 20.04.2023
• Subjects: Computer Science - Computer Vision and Pattern Recognition; Computer Science - Graphics
• DOI: 10.48550/arxiv.2304.10537

Neural radiance fields (NeRFs) enable novel view synthesis with unprecedented visual quality. However, to render photorealistic images, NeRFs require hundreds of deep multilayer perceptron (MLP) evaluations - for each pixel. This is prohibitively expensive and makes real-time rendering infeasible, even on powerful modern GPUs. In this paper, we propose a novel approach to distill and bake NeRFs into highly efficient mesh-based neural representations that are fully compatible with the massively parallel graphics rendering pipeline. We represent scenes as neural radiance features encoded on a two-layer duplex mesh, which effectively overcomes the inherent inaccuracies in 3D surface reconstruction by learning the aggregated radiance information from a reliable interval of ray-surface intersections. To exploit local geometric relationships of nearby pixels, we leverage screen-space convolutions instead of the MLPs used in NeRFs to achieve high-quality appearance. Finally, the performance of the whole framework is further boosted by a novel multi-view distillation optimization strategy. We demonstrate the effectiveness and superiority of our approach via extensive experiments on a range of standard datasets.