InfiniMotion：Mamba提升Transformer的记忆能力以实现任意长运动生成
InfiniMotion: Mamba Boosts Memory in Transformer for Arbitrary Long Motion Generation
AI总结：
针对现有文本到运动方法难以高效生成一致长序列的问题，提出InfiniMotion，通过带双向Mamba记忆的运动记忆Transformer在自回归框架下实现任意长运动生成，性能大幅提升。
摘要：
文本到运动生成在电影、游戏和机器人领域具有应用潜力，但现有方法通常优先关注短运动生成，难以有效生成运动长序列：（1）由于计算成本过高，现有方法难以将长运动序列作为单输入处理；（2）将长运动序列生成拆分为更短片段会导致过渡不一致，且需要插值或修复，缺乏对整个序列的建模。为解决这些挑战，我们提出InfiniMotion，一种在自回归框架内生成任意长度连续运动序列的方法。我们通过生成一段约80000帧的1小时连续人体运动，展现了其突破性能力。具体而言，我们引入了带有双向Mamba记忆的运动记忆Transformer（Motion Memory Transformer），增强Transformer的记忆能力，使其能在不消耗过多计算资源的情况下有效处理长运动序列。值得注意的是，我们的方法在FID指标上实现了超过30%的提升，演示时长是此前最优方法的6倍，展现了长运动生成领域的显著进展。项目网页：https://steve-zeyu-zhang.github.io/InfiniMotion/
英文摘要：
Text-to-motion generation holds potential for film, gaming, and robotics, yet current methods often prioritize short motion generation, making it challenging to produce long motion sequences effectively: (1) Current methods struggle to handle long motion sequences as a single input due to prohibitively high computational cost; (2) Breaking down the generation of long motion sequences into shorter segments can result in inconsistent transitions and requires interpolation or inpainting, which lacks entire sequence modeling. To solve these challenges, we propose InfiniMotion, a method that generates continuous motion sequences of arbitrary length within an autoregressive framework. We highlight its groundbreaking capability by generating a continuous 1-hour human motion with around 80,000 frames. Specifically, we introduce the Motion Memory Transformer with Bidirectional Mamba Memory, enhancing the transformer's memory to process long motion sequences effectively without overwhelming computational resources. Notably our method achieves over 30% improvement in FID and 6 times longer demonstration compared to previous state-of-the-art methods, showcasing significant advancements in long motion generation. See project webpage: https://steve-zeyu-zhang.github.io/InfiniMotion/
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。
