AvatarPointillist：自回归 4D 高斯化身化
AvatarPointillist: AutoRegressive 4D Gaussian Avatarization
用仅解码器 Transformer 自回归地逐点生成 3DGS 点云，并按主体复杂度自适应调整点密度，这一「序列化生成 3D 表示」的范式可迁移到其他需要可变分辨率三维表示的生成任务；同时以 AR 生成器的潜变量特征作为高斯解码器的条件，为两阶段生成管线提供了跨阶段特征复用的具体做法。
原文依据 a decoder-only Transformer that autoregressively generates a point cloud for 3D Gaussian Splatting
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出 AvatarPointillist，一个从单张肖像图像生成动态 4D 高斯化身的新框架。本方法的核心是一个仅解码器 Transformer，它以自回归方式生成用于 3D 高斯泼溅的点云。这种序列化方式支持精确且自适应的构建，可根据主体的复杂度动态调整点密度与点的总数。在点生成过程中，自回归模型还联合预测逐点的绑定信息，从而实现逼真的动画驱动。生成完成后，一个专用的高斯解码器将这些点转换为完整、可渲染的高斯属性。我们证明，以自回归生成器的潜变量特征作为解码器的条件，能够实现两个阶段之间的有效交互，并显著提升保真度。大量实验验证了 AvatarPointillist 能够产生高质量、照片级真实且可控的化身。我们相信这种自回归形式代表了化身生成的一种新范式，并将开源代码以启发后续研究。
英文原文摘要 1091 字符
We introduce AvatarPointillist, a novel framework for generating dynamic 4D Gaussian avatars from a single portrait image. At the core of our method is a decoder-only Transformer that autoregressively generates a point cloud for 3D Gaussian Splatting. This sequential approach allows for precise, adaptive construction, dynamically adjusting point density and the total number of points based on the subject's complexity. During point generation, the AR model also jointly predicts per-point binding information, enabling realistic animation. After generation, a dedicated Gaussian decoder converts the points into complete, renderable Gaussian attributes. We demonstrate that conditioning the decoder on the latent features from the AR generator enables effective interaction between stages and markedly improves fidelity. Extensive experiments validate that AvatarPointillist produces high-quality, photorealistic, and controllable avatars. We believe this autoregressive formulation represents a new paradigm for avatar generation, and we will release our code to inspire future research.