nerfies.github.io (a) casual capture (b) input images (c) nerfie novel views (d) novel view depth Figure 1: We reconstruct photo-realistic nerfies from a user casually waving a mobile phone (a). Our system uses selfie photos/videos (b) to produce a free-viewpoint representation with renderings (c) and accurate scene geometry (d). Please see supplementary for video results.
We present a learning-based method for synthesizing novel views of complex outdoor scenes using only unstructured collections of in-the-wild photographs. We build on neural radiance fields (NeRF), which uses the weights of a multilayer perceptron to implicitly model the volumetric density and color of a scene. While NeRF works well on images of static subjects captured under controlled settings, it is incapable of modeling many ubiquitous, real-world phenomena in uncontrolled images, such as variable illumination or transient occluders. In this work, we introduce a series of extensions to NeRF to address these issues, thereby allowing for accurate reconstructions from unstructured image collections taken from the internet. We apply our system, which we dub NeRF-W, to internet photo collections of famous landmarks, thereby producing photorealistic, spatially consistent scene representations despite unknown and confounding factors, resulting in significant improvement over the state of the art.
Figure 1: This state-of-the-art report discusses a large variety of neural rendering methods which enable applications such as novel-view synthesis of static and dynamic scenes, generative modeling of objects, and scene relighting. See Section 4 for more details on the various methods. Images adapted from [
Figure 1: Our method obtains a high-quality 3D reconstruction from an RGB-D input sequence by training a multi-layer perceptron. The core idea is to reformulate the neural radiance field definition in NeRF [37], and replace it with a differentiable rendering formulation based on signed distance fields which is specifically tailored to geometry reconstruction.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.