DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación
Artículo científico (JournalArticle)
InstantGeoAvatar: Effective Geometry and Appearance Modeling of Animatable Avatars from Monocular Video
We present InstantGeoAvatar, a method for efficient and effective learning from monocular video of detailed 3D geometry and appearance of animatable implicit human avatars. Our key observation is that the optimization of a hash grid encoding to represent a signed distance function (SDF) of the human subject is fraught with instabilities and bad local minima. We thus propose a principled geometry-aware SDF regularization scheme that seamlessly fits into the volume rendering pipeline and adds negligible computational overhead. Our regularization scheme significantly outperforms previous approaches for training SDFs on hash grids. We obtain competitive results in geometry reconstruction and novel view synthesis in as little as five minutes of training time, a significant reduction from the several hours required by previous work. InstantGeoAvatar represents a significant leap forward towards achieving interactive reconstruction of virtual avatars., This work has been supported by the project MOHUCO PID2020-120049RB-I00 funded by MCIU/AEI/10.13039/501100011033, and by the project GRAVATAR PID2023-151184OB-I00 funded by MCIU/AEI/10.13039/501100011033 and by ERDF, UE., Peer reviewed
Proyecto:
AEI / Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020 PID2020-120049RB-I00
,
AEI / Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023 PID2023-151184OB-I00
DOI: http://hdl.handle.net/10261/387911, https://api.elsevier.com/content/abstract/scopus_id/85212964933
HANDLE: http://hdl.handle.net/10261/387911, https://api.elsevier.com/content/abstract/scopus_id/85212964933