InvertAvatar：用于通用头部虚拟形象的增量式GAN反演
InvertAvatar: Incremental GAN Inversion for Generalized Head Avatars
AI总结：
提出增量式3D GAN反演框架InvertAvatar，利用多帧输入提升头部虚拟形象重建保真度，通过可动画3D GAN先验和神经纹理编码器增强表情控制与细节，实现一次性及少样本动画任务的最先进性能。
摘要：
虽然高保真度和效率是创建数字头部虚拟形象的核心，但近期依赖2D或3D生成模型的方法常常面临形状扭曲、表情不准确和身份闪烁等限制。此外，现有的一次性反演技术未能充分利用多张输入图像进行细节特征提取。我们提出了一种新颖的框架——增量式3D GAN反演，通过一种旨在提高多帧保真度的算法来增强虚拟形象重建性能，从而获得与帧数成正比的重建质量提升。我们的方法引入了一种独特的可动画3D GAN先验，包含两项关键修改以增强表情可控性，以及一个创新的神经纹理编码器，该编码器基于UV参数化对纹理特征空间进行分类。与传统技术不同，我们的架构强调像素对齐的图像到图像转换，减少了学习观测空间与规范空间之间对应关系的需求。此外，我们整合了基于ConvGRU的循环网络，用于多帧时间数据聚合，增强了几何和纹理细节的重建。所提出的范式在一次性及少样本虚拟形象动画任务中展示了最先进的性能。代码将可在 https://github.com/XChenZ/invertAvatar 获取。
英文摘要：
While high fidelity and efficiency are central to the creation of digital head avatars, recent methods relying on 2D or 3D generative models often experience limitations such as shape distortion, expression inaccuracy, and identity flickering. Additionally, existing one-shot inversion techniques fail to fully leverage multiple input images for detailed feature extraction. We propose a novel framework, \textbf{Incremental 3D GAN Inversion}, that enhances avatar reconstruction performance using an algorithm designed to increase the fidelity from multiple frames, resulting in improved reconstruction quality proportional to frame count. Our method introduces a unique animatable 3D GAN prior with two crucial modifications for enhanced expression controllability alongside an innovative neural texture encoder that categorizes texture feature spaces based on UV parameterization. Differentiating from traditional techniques, our architecture emphasizes pixel-aligned image-to-image translation, mitigating the need to learn correspondences between observation and canonical spaces. Furthermore, we incorporate ConvGRU-based recurrent networks for temporal data aggregation from multiple frames, boosting geometry and texture detail reconstruction. The proposed paradigm demonstrates state-of-the-art performance on one-shot and few-shot avatar animation tasks. Code will be available at https://github.com/XChenZ/invertAvatar.
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。
