SpatialAvatar-0:多阶段重建的高质量4D头部头像
阅读原文· arxiv.orgSpatialAvatar-0 提出基于 FLAME-mesh 约束的高斯表示,结合前馈生成器与 10K 迭代布局保持的逐主体精化循环。前馈阶段采用无参数 K 源均值池化及单目-时序到多视图-空间两阶段调度;精化阶段冻结 FLAME 绑定与高斯数量,以三组件抗尖峰正则化替代密集化。在 VFHQ/HDTF 跨域零样本测试中,PSNR 超越领域内领先模型 GAGAvatar 1.5 dB;在 SplattingAvatar 单目基准上,所有指标均领先,PSNR 超越 300K 迭代的 GeoAvatar 1.3 dB,且逐主体调度周期比常见 SOTA 基线快 60 倍。
High-quality 4D head avatars from one or a few source portraits are central to telepresence, AR/VR, and digital-human interaction. 3D Gaussian Splatting (3DGS) has emerged as the dominant representation, with two complementary regimes (generalizable feed-forward predictors and per-subject refiners) maturing in parallel. However, existing feed-forward predictors are trained on a single dataset family with a hard-coded source count, inheriting the corresponding domain bias. Per-subject refiners require 300K--600K iterations and rely on adaptive densification that destroys upstream Gaussian layouts, preventing the two regimes from sharing a representation end-to-end. To bridge both regimes we propose SpatialAvatar-0 on a shared FLAME-mesh-bound Gaussian representation: a feed-forward generator with a parameter-free K-source mean-pool and a monocular-temporal to multi-view-spatial two-phase schedule that anchors against identity-prior collapse onto the smaller multi-view set. We further introduce a 10K-iter layout-preserving per-subject refinement loop that freezes the FLAME-binding and Gaussian count and replaces densification with a three-component anti-spike regularization. On VFHQ/HDTF cross-domain zero-shot we surpass the in-domain leader GAGAvatar by +1.5 dB PSNR despite never training on either test domain, and on the SplattingAvatar monocular benchmark we lead every reported metric, surpassing the 300K-iter GeoAvatar by +1.3 dB PSNR at up to 60x shorter per-subject schedule than common SOTA baselines. Website: https://spatialwalk.github.io/SpatialAvatar-0.