用事件相机拍视频，AI 能补出高清慢动作
事件相机只记录像素亮度的变化，不拍完整帧，数据稀疏但极快（微秒级）。传统方法要么模糊，要么生成几秒就崩。这篇把预训练的视频扩散模型（类似 Stable Video Diffusion）微调后，同时做三件事：从事件流重建视频、预测未来帧、插帧。关键创新是「自回归展开」和「自适应上下文切换」——把长视频切成短段，每段只依赖前一段的少量信息，避免误差累积；插帧时用「重编码对齐」保证前后一致。在真实数据集上，它生成的视频比现有方法更清晰、更连贯，甚至能零样本泛化到没见过的场景。它不是你能明天用的工具，但意味着未来用事件相机（比如高速运动、暗光环境）拍的东西，AI 能补出流畅的高清视频，这对自动驾驶、机器人、高速摄影是实打实的进步。
📄 原文摘要(英文)
Recovering high-quality video from sparse event streams is a challenging task. Regression methods often blur textures, while existing generative models struggle with long-term stability. We propose LongE2V, a novel approach that leverages pre-trained video diffusion priors to jointly handle event-based video reconstruction, prediction, and frame interpolation. By fine-tuning a foundational video model, our approach achieves high data efficiency and superior perceptual quality. We introduce Autoregressive Unrolling and Adaptive Context Switching to mitigate temporal drift in extremely long sequences. We also propose Reencoding Alignment with Cross Residual Correction to ensure precise bidirectional consistency during frame interpolation. Furthermore, Event Voxel Density Augmentation ensures robustness across varying sensor resolutions. Extensive experiments on real-world benchmarks demonstrate that LongE2V outperforms state-of-the-art methods across all three tasks, exhibiting exceptional temporal coherence and zero-shot generalization. Project page: https://cdfan0627.github.io/LongE2V-page/