SpatialAvatar-0: 高质量4D头部虚拟形象的多阶段重建
SpatialAvatar-0: High-Quality 4D Head Avatar with Multi-Stage Reconstruction
AI总结：
提出基于FLAME网格绑定高斯表示的多阶段框架，通过前馈生成器和10K迭代布局保持微调，实现跨域零样本和单目基准领先性能。
摘要：
高质量4D头部虚拟形象（来自一张或少量源肖像）是远程呈现、AR/VR和数字人交互的核心。3D高斯泼溅（3DGS）已成为主导表示，两个互补范式（可泛化的前馈预测器和逐主体精炼器）并行成熟。然而，现有前馈预测器在单一数据集族上训练，具有硬编码的源数量，继承了相应的领域偏差。逐主体精炼器需要30万至60万次迭代，并依赖自适应致密化，这会破坏上游高斯布局，导致两个范式无法端到端共享表示。为桥接两个范式，我们提出SpatialAvatar-0，基于共享的FLAME网格绑定高斯表示：一个前馈生成器，具有无参数的K源均值池化，以及一个从单目时序到多视角空间的两阶段调度，防止身份先验在小多视角集上坍缩。我们进一步引入一个10K迭代的布局保持逐主体精炼循环，冻结FLAME绑定和高斯数量，并用三分量抗尖峰正则化替代致密化。在VFHQ/HDTF跨域零样本上，我们超越域内领先者GAGAvatar +1.5 dB PSNR，尽管从未在任一测试域上训练；在SplattingAvatar单目基准上，我们领先所有报告指标，超越30万次迭代的GeoAvatar +1.3 dB PSNR，且逐主体调度比常见SOTA基线短至60倍。网站：https://spatialwalk.github.io/SpatialAvatar-0。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
