用小数据微调开源视频模型，实现影视级场景生成。
Fine-Tuning Open Video Generators for Cinematic Scene Synthesis: A Small-Data Pipeline with LoRA and Wan2.1 I2V
- 分两阶段：先用LoRA学风格，再生成连贯视频帧
- 在单张显卡上数小时内完成微调，生成720p流畅视频
- 适合影视制作团队快速定制视觉风格
我们提出一种实用的微调流程，用于将开源视频扩散模型适配至电视与电影制作中的影视场景合成任务，仅需少量数据。该两阶段流程将视觉风格学习与运动生成解耦：第一阶段在Wan2.1 I2V-14B模型的交叉注意力层中引入低秩适配（LoRA）模块，利用来自Ay Yapim历史剧《El Turco》的短片段数据集，实现高效领域迁移，可在单张GPU上数小时内完成；第二阶段使用微调后的模型生成风格一致的关键帧，保留服装、灯光与色彩分级特征，并通过模型视频解码器时序扩展为720p连贯序列。进一步采用轻量级并行化与序列分块策略，在不降低质量的前提下加速推理。基于FVD、CLIP-SIM和LPIPS指标的定量评估，结合小规模专家用户研究，证明其在影视真实感与时间稳定性方面显著优于基线模型。完整训练与推理流程已开源，支持跨影视领域的可复现与适应。
原文摘要 · Abstract (English)
We present a practical pipeline for fine-tuning open-source video diffusion transformers to synthesize cinematic scenes for television and film production from small datasets. The proposed two-stage process decouples visual style learning from motion generation. In the first stage, Low-Rank Adaptation (LoRA) modules are integrated into the cross-attention layers of the Wan2.1 I2V-14B model to adapt its visual representations using a compact dataset of short clips from Ay Yapim's historical television film El Turco. This enables efficient domain transfer within hours on a single GPU. In the second stage, the fine-tuned model produces stylistically consistent keyframes that preserve costume, lighting, and color grading, which are then temporally expanded into coherent 720p sequences through the model's video decoder. We further apply lightweight parallelization and sequence partitioning strategies to accelerate inference without quality degradation. Quantitative and qualitative evaluations using FVD, CLIP-SIM, and LPIPS metrics, supported by a small expert user study, demonstrate measurable improvements in cinematic fidelity and temporal stability over the base model. The complete training and inference pipeline is released to support reproducibility and adaptation across cinematic domains.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。