闪电快的基于缓存的并行去噪预测：加速说唱头生成
Lightning Fast Caching-based Parallel Denoising Prediction for Accelerating Talking Head Generation
AI总结：
本文提出闪电快的基于缓存的并行去噪预测和解耦前景注意力方法，通过优化缓存和注意力计算，加速说唱头生成的推理过程，同时保持视频质量。
摘要：
基于扩散的说唱头模型能够生成高质量、逼真的视频，但存在推理速度慢的问题，限制了实际应用。现有的通用扩散模型加速方法无法利用说唱头生成中特有的时间和空间冗余性。在本文中，我们提出了一种任务特定的框架，通过两个关键创新来解决这些低效问题。首先，我们引入了闪电快的基于缓存的并行去噪预测（LightningCP），通过缓存静态特征来绕过推理时间中大部分模型层。我们还利用缓存的特征和估计的噪声潜在特征作为输入，实现并行预测，高效地绕过了顺序采样。其次，我们提出了解耦前景注意力（DFA），利用说唱头视频中的空间解耦性，将注意力限制在动态前景区域。此外，我们移除某些层中的参考特征以进一步提升速度。广泛的实验表明，我们的框架显著提高了推理速度，同时保持视频质量。
英文摘要：
Diffusion-based talking head models generate high-quality, photorealistic videos but suffer from slow inference, limiting practical applications. Existing acceleration methods for general diffusion models fail to exploit the temporal and spatial redundancies unique to talking head generation. In this paper, we propose a task-specific framework addressing these inefficiencies through two key innovations. First, we introduce Lightning-fast Caching-based Parallel denoising prediction (LightningCP), caching static features to bypass most model layers in inference time. We also enable parallel prediction using cached features and estimated noisy latents as inputs, efficiently bypassing sequential sampling. Second, we propose Decoupled Foreground Attention (DFA) to further accelerate attention computations, exploiting the spatial decoupling in talking head videos to restrict attention to dynamic foreground regions. Additionally, we remove reference features in certain layers to bring extra speedup. Extensive experiments demonstrate that our framework significantly improves inference speed while preserving video quality.
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。
