Hallucinated Neural Radiance Fields in the Wild
This addresses the challenge of time-varying appearance synthesis in 3D scene reconstruction for applications like virtual tourism, but it is incremental as it builds on existing NeRF methods.
The paper tackles the problem of generating Neural Radiance Fields (NeRF) at different times of day from tourism images, presenting Ha-NeRF, which hallucinates appearances and handles occlusions to render view-consistent, occlusion-free images.
Neural Radiance Fields (NeRF) has recently gained popularity for its impressive novel view synthesis ability. This paper studies the problem of hallucinated NeRF: i.e., recovering a realistic NeRF at a different time of day from a group of tourism images. Existing solutions adopt NeRF with a controllable appearance embedding to render novel views under various conditions, but they cannot render view-consistent images with an unseen appearance. To solve this problem, we present an end-to-end framework for constructing a hallucinated NeRF, dubbed as Ha-NeRF. Specifically, we propose an appearance hallucination module to handle time-varying appearances and transfer them to novel views. Considering the complex occlusions of tourism images, we introduce an anti-occlusion module to decompose the static subjects for visibility accurately. Experimental results on synthetic data and real tourism photo collections demonstrate that our method can hallucinate the desired appearances and render occlusion-free images from different views. The project and supplementary materials are available at https://rover-xingyu.github.io/Ha-NeRF/.