ChopGrad：通过截断反向传播实现基于像素的潜在视频扩散模型
ChopGrad: Pixel-Wise Losses for Latent Video Diffusion via Truncated Backpropagation
AI总结：
本文提出ChopGrad，通过截断反向传播减少视频生成的内存消耗，实现高效的像素级损失微调，适用于视频超分辨率、修复和增强等任务。
摘要：
最近的视频扩散模型通过递归帧处理实现高质量生成，但这种递归机制导致像素域训练内存成本高昂。本文提出ChopGrad，通过限制梯度计算到局部帧窗口，保持全局一致性，理论分析显示其能高效微调。ChopGrad将训练内存从与视频帧数线性增长降至常量，且在多个条件视频生成任务中表现优异。
英文摘要：
Recent video diffusion models achieve high-quality generation through recurrent frame processing where each frame generation depends on previous frames. However, this recurrent mechanism means that training such models in the pixel domain incurs prohibitive memory costs, as activations accumulate across the entire video sequence. This fundamental limitation also makes fine-tuning these models with pixel-wise losses computationally intractable for long or high-resolution videos. This paper introduces ChopGrad, a truncated backpropagation scheme for video decoding, limiting gradient computation to local frame windows while maintaining global consistency. We provide a theoretical analysis of this approximation and show that it enables efficient fine-tuning with frame-wise losses. ChopGrad reduces training memory from scaling linearly with the number of video frames (full backpropagation) to constant memory, and compares favorably to existing state-of-the-art video diffusion models across a suite of conditional video generation tasks with pixel-wise losses, including video super-resolution, video inpainting, video enhancement of neural-rendered scenes, and controlled driving video generation.
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。
