SKIP: 用于高效具身世界模型的稀疏关键帧插值范式
SKIP: Sparse Keyframe Interpolation Paradigm for Efficient Embodied World Models
AI总结：
提出稀疏关键帧插值范式(SKIP)，通过识别任务相关关键帧并仅生成这些帧，再基于机器人动作插值缺失帧，实现高效视频生成，在LIBERO上速度提升4.16倍，FVD降低89%，且生成视频作为训练数据时策略性能下降极小。
摘要：
具身世界模型通过预测机器人动作如何影响周围场景，已成为机器人学中一种有前景的范式。然而，在像素空间中进行 rollout 推理在计算上仍然昂贵，因为长时程操作视频通常必须逐帧生成。这种成本不能通过不加区分地丢弃帧来轻易降低，因为下游策略依赖于对稀疏任务相关事件（如接近、接触、抓取和释放）的完整保留。为了解决这一挑战，我们提出了稀疏关键帧插值范式（SKIP），这是一种事件保留的稀疏到密集框架，避免了密集的逐帧生成。SKIP 首先通过利用机器人感知的多模态特征来识别任务相关的关键帧。然后，它仅用稀疏视频扩散模型合成这些关键帧。一个学习到的间隙预测器和一个动作条件插值器随后根据机器人动作重建缺失的间隔。在 LIBERO 上，SKIP 生成密集 rollouts 的速度比密集基线快 4.16 倍，同时提高了视觉保真度并将聚合 FVD 降低了 89.0%。重要的是，SKIP 生成的视频是有效的策略训练数据。即使它们完全替代真实演示，π_{0.5} 的成功率在 LIBERO 模拟中仅下降 1.3 个百分点，在真实机器人上下降 6.7 个百分点，而完全密集的逐帧生成则下降 48 到 58 个百分点。
英文摘要：
Embodied world models have emerged as a promising paradigm in robotics by predicting how robot actions affect the surrounding scene. However, the rollout inference remains computationally expensive in pixel space, as long-horizon manipulation videos typically have to be generated frame by frame. This cost cannot be easily reduced by indiscriminately dropping frames, since downstream policies rely on complete preservation of sparse task-relevant events such as approach, contact, grasp, and release. To address this challenge, we propose Sparse Keyframe Interpolation Paradigm (SKIP), an event-preserving sparse-to-dense framework that avoids dense frame-by-frame generation. SKIP first identifies task-relevant keyframes by leveraging robot-aware multimodal features. It then synthesizes only these keyframes with a sparse video diffusion model. A learned gap predictor and an action-conditioned interpolator subsequently reconstruct the missing intervals according to the robot actions. On LIBERO, SKIP generates dense rollouts $4.16\times$ faster than a dense baseline while improving visual fidelity and reducing aggregate FVD by $89.0\%$. Importantly, SKIP-generated videos are effective policy-training data. Even when they fully replace real demonstrations, $π_{0.5}$ success drops only $1.3$ pp in LIBERO simulation and $6.7$ pp on the real robot, whereas fully dense frame-by-frame generation collapses by $48$ to $58$ pp.
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。
