LEO：用于人体视频合成的生成式潜在图像动画器
LEO: Generative Latent Image Animator for Human Video Synthesis
AI总结：
提出LEO框架，通过光流图和LMDM实现外观与运动解耦，显著提升人体视频合成的时空一致性，并支持无限长视频合成与内容保持编辑。
摘要：
时空一致性是合成高质量视频的主要挑战，尤其是在合成包含丰富全局和局部形变的人体视频时。为了解决这一挑战，以往的方法在生成过程中采用了不同的特征来表示外观和运动。然而，由于缺乏严格的机制来保证这种解耦，将运动与外观分离仍然具有挑战性，导致空间失真和时间抖动，破坏了时空一致性。受此启发，我们在此提出LEO，一种用于人体视频合成的新型框架，重点关注时空一致性。我们的核心思想是在生成过程中将运动表示为一系列光流图，从而本质上将运动与外观隔离开来。我们通过基于光流的图像动画器和Latent Motion Diffusion Model (LMDM)来实现这一思想。前者将运动编码空间与光流图空间连接起来，以扭曲和修复的方式合成视频帧。LMDM通过合成运动编码序列来学习捕捉训练数据中的运动先验。广泛的定量和定性分析表明，LEO在TaichiHD、FaceForensics和CelebV-HQ数据集上，相比以往方法显著提升了人体视频的一致性合成。此外，LEO中外观和运动的有效解耦还支持两项额外任务，即无限长度人体视频合成以及内容保持的视频编辑。
英文摘要：
Spatio-temporal coherency is a major challenge in synthesizing high quality videos, particularly in synthesizing human videos that contain rich global and local deformations. To resolve this challenge, previous approaches have resorted to different features in the generation process aimed at representing appearance and motion. However, in the absence of strict mechanisms to guarantee such disentanglement, a separation of motion from appearance has remained challenging, resulting in spatial distortions and temporal jittering that break the spatio-temporal coherency. Motivated by this, we here propose LEO, a novel framework for human video synthesis, placing emphasis on spatio-temporal coherency. Our key idea is to represent motion as a sequence of flow maps in the generation process, which inherently isolate motion from appearance. We implement this idea via a flow-based image animator and a Latent Motion Diffusion Model (LMDM). The former bridges a space of motion codes with the space of flow maps, and synthesizes video frames in a warp-and-inpaint manner. LMDM learns to capture motion prior in the training data by synthesizing sequences of motion codes. Extensive quantitative and qualitative analysis suggests that LEO significantly improves coherent synthesis of human videos over previous methods on the datasets TaichiHD, FaceForensics and CelebV-HQ. In addition, the effective disentanglement of appearance and motion in LEO allows for two additional tasks, namely infinite-length human video synthesis, as well as content-preserving video editing.
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。
