Fine-Tuning Open Video Generators for Cinematic Scene Synthesis: A Small-Data Pipeline with LoRA and Wan2.1 I2V

TL;DR

This paper introduces a practical, efficient pipeline for fine-tuning open-source video diffusion models to generate cinematic scenes from small datasets, combining LoRA adaptation and a two-stage process for style and motion synthesis.

Contribution

The paper presents a novel two-stage fine-tuning pipeline using LoRA modules and a video decoder, enabling cinematic scene synthesis from small datasets with high fidelity and temporal coherence.

Findings

Improved cinematic fidelity and temporal stability demonstrated by quantitative metrics.

Efficient domain transfer achieved within hours on a single GPU.

Pipeline supports reproducibility and adaptation across cinematic domains.

Abstract

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.