用约束反演稳定稀疏监督下的虚拟人编辑，防身份泄露和抖动。
Information-Regularized Constrained Inversion for Stable Avatar Editing from Sparse Supervision
- 在结构化潜空间中做受约束的反演，限定更新到局部编辑子空间。
- 通过优化条件目标构建信息矩阵，预测并稳定编辑结果。
- 适合资源受限下精细虚拟人编辑，尤其适用于少样本场景。
虚拟人编辑通常依赖少量标注关键帧，但直接拟合这些编辑常导致身份泄露和姿态相关的时序闪烁。我们指出此类失败源于病态反演：可用编辑约束不足以确定目标编辑的潜在方向。为此，提出一种基于条件引导的编辑重建框架，将编辑建模为结构化虚拟人潜空间中的约束反演，限制更新仅在低维、部件特定的编辑子空间内，防止意外身份变化。关键在于，通过优化一个源自完整解码与渲染流水线局部线性化的条件目标，构建编辑子空间的信息矩阵，其谱可预测稳定性，并用于帧权重调整或关键帧激活。该方法仅操作小规模子空间矩阵，可通过黑塞-向量乘积等高效实现，在有限编辑监督下显著提升稳定性。
原文摘要 · Abstract (English)
Editing animatable human avatars typically relies on sparse supervision, often a few edited keyframes, yet naively fitting a reconstructed avatar to these edits frequently causes identity leakage and pose-dependent temporal flicker. We argue that these failures are best understood as an ill-conditioned inversion: the available edited constraints do not sufficiently determine the latent directions responsible for the intended edit. We propose a conditioning-guided edited reconstruction framework that performs editing as a constrained inversion in a structured avatar latent space, restricting updates to a low-dimensional, part-specific edit subspace to prevent unintended identity changes. Crucially, we design the editing constraints during inversion by optimizing a conditioning objective derived from a local linearization of the full decoding-and-rendering pipeline, yielding an edit-subspace information matrix whose spectrum predicts stability and drives frame reweighting / keyframe activation. The resulting method operates on small subspace matrices and can be implemented efficiently (e.g., via Hessian-vector products), and improves stability under limited edited supervision.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。