Stylizing Sparse-View 3D Scenes with Hierarchical Neural Representation

arXiv:2404.05236v1 Announce Type: new
Abstract: Recently, a surge of 3D style transfer methods has been proposed that leverage the scene reconstruction power of a pre-trained neural radiance field (NeRF). To successfully stylize a scene this way, one must first reconstruct a photo-realistic radiance field from collected images of the scene. However, when only sparse input views are available, pre-trained few-shot NeRFs often suffer from high-frequency artifacts, which are generated as a by-product of high-frequency details for improving reconstruction quality. Is …

3d scenes abstract arxiv cs.cv cs.gr hierarchical however images nerf neural radiance field photo power representation style style transfer stylize transfer type view