We present Split-then-Merge (StM), a novel framework designed to enhance control in generative video composition and address its data scarcity problem. Unlike conventional methods relying on annotated datasets or handcrafted rules, StM splits a large corpus of unlabeled videos into dynamic foreground and background layers, then self-composes them to learn how dynamic subjects interact with diverse scenes. This process enables the model to learn the complex compositional dynamics required for realistic video generation. StM introduces a novel transformation-aware training pipeline that utilizes a multi-layer fusion and augmentation to achieve affordance-aware composition, alongside an identity-preservation loss that maintains foreground fidelity during blending. Experiments show StM outperforms SoTA methods in both quantitative benchmarks and in humans/VLLM-based qualitative evaluations. More details are available at our project page: https://split-then-merge.github.io
@article{arxiv.2511.20809,
title = {Layer-Aware Video Composition via Split-then-Merge},
author = {Ozgur Kara and Yujia Chen and Ming-Hsuan Yang and James M. Rehg and Wen-Sheng Chu and Du Tran},
journal= {arXiv preprint arXiv:2511.20809},
year = {2025}
}