GaussianAvatar-Editor：逼真可動的高斯頭像編輯器

GaussianAvatar-Editor: Photorealistic Animatable Gaussian Head Avatar Editor

January 17, 2025

作者: Xiangyue Liu, Kunming Luo, Heng Li, Qi Zhang, Yuan Liu, Li Yi, Ping Tan

cs.AI

摘要

我們介紹了 GaussianAvatar-Editor，這是一個創新的框架，用於基於文本進行可動態控制的高斯頭像編輯，可以完全控制表情、姿勢和視角。與靜態 3D 高斯編輯不同，編輯可動態的 4D 高斯頭像面臨與運動遮擋和空間-時間不一致性相關的挑戰。為了應對這些問題，我們提出了加權 Alpha 混合方程（WABE）。該函數增強可見高斯的混合權重，同時抑制對不可見高斯的影響，有效處理編輯過程中的運動遮擋。此外，為了提高編輯質量並確保 4D 一致性，我們將條件對抗學習融入編輯過程中。這一策略有助於改進編輯結果並在整個動畫過程中保持一致性。通過整合這些方法，我們的 GaussianAvatar-Editor 在可動態的 4D 高斯編輯中實現了逼真且一致的結果。我們在各種主題上進行了全面的實驗，以驗證我們提出的技術的有效性，這證明了我們方法的優越性。更多結果和代碼可在以下項目鏈接中找到：[項目鏈接](https://xiangyueliu.github.io/GaussianAvatar-Editor/)。

English

We introduce GaussianAvatar-Editor, an innovative framework for text-driven editing of animatable Gaussian head avatars that can be fully controlled in expression, pose, and viewpoint. Unlike static 3D Gaussian editing, editing animatable 4D Gaussian avatars presents challenges related to motion occlusion and spatial-temporal inconsistency. To address these issues, we propose the Weighted Alpha Blending Equation (WABE). This function enhances the blending weight of visible Gaussians while suppressing the influence on non-visible Gaussians, effectively handling motion occlusion during editing. Furthermore, to improve editing quality and ensure 4D consistency, we incorporate conditional adversarial learning into the editing process. This strategy helps to refine the edited results and maintain consistency throughout the animation. By integrating these methods, our GaussianAvatar-Editor achieves photorealistic and consistent results in animatable 4D Gaussian editing. We conduct comprehensive experiments across various subjects to validate the effectiveness of our proposed techniques, which demonstrates the superiority of our approach over existing methods. More results and code are available at: [Project Link](https://xiangyueliu.github.io/GaussianAvatar-Editor/).