发表机构
EverEx; Yonsei University; Korea University（EverEx; 延世大学; 高丽大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
FlashRender是一种少步生成渲染框架，通过RETA、MeanFlow及在线策略流图蒸馏互补技术，实现25倍低采样成本下媲美多步基线的视频质量与几何一致性，且相机可控性更优。
AI 中文摘要
我们提出了FlashRender，这是一种少步生成渲染框架，可在数秒内沿目标相机轨迹重绘源视频。我们发现，依赖采样步数的相机控制是现有多步生成渲染模型中离散化误差的突出表现，且解决该不一致性可大幅降低去噪轨迹曲率，为后续的步数蒸馏提供便利。为此，我们引入了表示变换与对齐（RETA），该方法将源视频的隐藏表示与来自冻结视觉几何模型的目标视频特征进行对齐，直接在源视频流中编码几何变换，实现了与采样步数一致的相机控制。随后，我们在RETA诱导的低曲率去噪轨迹上，通过MeanFlow目标对模型进行微调，使模型能更有效地解决离散化误差问题。最后，我们应用在线策略流图蒸馏，以修正固定少步采样下的自推出误差。大量实验表明，RETA、MeanFlow及在线策略流图蒸馏在少步生成渲染中发挥互补作用，三者结合使我们的方法在采样成本降低25倍的情况下，视频质量与几何一致性可媲美多步基线，且在分布外目标相机轨迹下仍能实现更优的相机可控性。
英文摘要
We present FlashRender, a few-step generative rendering framework that retakes a source video along a target camera trajectory in seconds. We identify sampling-step-dependent camera control as a prominent manifestation of discretization error in existing multi-step generative rendering models and show that resolving this inconsistency substantially lowers denoising trajectory curvature, facilitating subsequent step distillation. To this end, we introduce Representation Transformation and Alignment (RETA), which aligns hidden source-video representations with target-video features from a frozen visual geometry model. This directly encodes the geometric transformation within the source-video stream, enabling sampling-step-consistent camera control. We then fine-tune the model with the MeanFlow objective on the lower-curvature denoising trajectory induced by RETA, allowing the model to more effectively address discretization error. Finally, we apply on-policy flow map distillation to correct self-rollout errors under fixed few-step sampling. Extensive experiments show that RETA, MeanFlow, and on-policy flow map distillation play complementary roles in few-step generative rendering. Together, they enable our approach to match multi-step baselines in video quality and geometric consistency at 25x lower sampling cost while achieving superior camera controllability, even under out-of-distribution target camera trajectories.
CommentsProject page: https://byeongjun-park.github.io/FlashRender/
