Animating NeRFs from Texture Space: A Framework for Pose-Dependent Rendering of Human Performances

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Knoll, Paul, Morgenstern, Wieland, Hilsmann, Anna, Eisert, Peter
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911802297679872
author Knoll, Paul
Morgenstern, Wieland
Hilsmann, Anna
Eisert, Peter
author_facet Knoll, Paul
Morgenstern, Wieland
Hilsmann, Anna
Eisert, Peter
contents Creating high-quality controllable 3D human models from multi-view RGB videos poses a significant challenge. Neural radiance fields (NeRFs) have demonstrated remarkable quality in reconstructing and free-viewpoint rendering of static as well as dynamic scenes. The extension to a controllable synthesis of dynamic human performances poses an exciting research question. In this paper, we introduce a novel NeRF-based framework for pose-dependent rendering of human performances. In our approach, the radiance field is warped around an SMPL body mesh, thereby creating a new surface-aligned representation. Our representation can be animated through skeletal joint parameters that are provided to the NeRF in addition to the viewpoint for pose dependent appearances. To achieve this, our representation includes the corresponding 2D UV coordinates on the mesh texture map and the distance between the query point and the mesh. To enable efficient learning despite mapping ambiguities and random visual variations, we introduce a novel remapping process that refines the mapped coordinates. Experiments demonstrate that our approach results in high-quality renderings for novel-view and novel-pose synthesis.
format Preprint
id arxiv_https___arxiv_org_abs_2311_03140
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Animating NeRFs from Texture Space: A Framework for Pose-Dependent Rendering of Human Performances
Knoll, Paul
Morgenstern, Wieland
Hilsmann, Anna
Eisert, Peter
Computer Vision and Pattern Recognition
Creating high-quality controllable 3D human models from multi-view RGB videos poses a significant challenge. Neural radiance fields (NeRFs) have demonstrated remarkable quality in reconstructing and free-viewpoint rendering of static as well as dynamic scenes. The extension to a controllable synthesis of dynamic human performances poses an exciting research question. In this paper, we introduce a novel NeRF-based framework for pose-dependent rendering of human performances. In our approach, the radiance field is warped around an SMPL body mesh, thereby creating a new surface-aligned representation. Our representation can be animated through skeletal joint parameters that are provided to the NeRF in addition to the viewpoint for pose dependent appearances. To achieve this, our representation includes the corresponding 2D UV coordinates on the mesh texture map and the distance between the query point and the mesh. To enable efficient learning despite mapping ambiguities and random visual variations, we introduce a novel remapping process that refines the mapped coordinates. Experiments demonstrate that our approach results in high-quality renderings for novel-view and novel-pose synthesis.
title Animating NeRFs from Texture Space: A Framework for Pose-Dependent Rendering of Human Performances
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.03140