ONE-SHOT: Compositional Human-Environment Video Synthesis via Spatial-Decoupled Motion Injection and Hybrid Context Integration

2026-04-01

AI Summary

Existing video generation methods struggle to enable independent, fine-grained editing of human subjects and their environments, often facing a trade-off between reliance on 3D preprocessing and control flexibility. This work proposes a parameter-efficient, composable framework for human–environment video generation that decouples human motion from scene signals via a canonical-space injection mechanism and introduces Dynamic-Grounded-RoPE positional encoding, which operates without 3D alignment. By integrating spatially disentangled cross-attention with a hybrid context fusion strategy, the method supports minute-long video synthesis while significantly outperforming current approaches in both structural control accuracy and creative diversity, thereby achieving efficient and flexible co-generation of humans and environments.

Abstract

Recent advances in Video Foundation Models (VFMs) have revolutionized human-centric video synthesis, yet fine-grained and independent editing of subjects and scenes remains a critical challenge. Recent attempts to incorporate richer environment control through rigid 3D geometric compositions often encounter a stark trade-off between precise control and generative flexibility. Furthermore, the heavy 3D pre-processing still limits practical scalability. In this paper, we propose ONE-SHOT, a parameter-efficient framework for compositional human-environment video generation. Our key insight is to factorize the generative process into disentangled signals. Specifically, we introduce a canonical-space injection mechanism that decouples human dynamics from environmental cues via cross-attention. We also propose Dynamic-Grounded-RoPE, a novel positional embedding strategy that establishes spatial correspondences between disparate spatial domains without any heuristic 3D alignments. To support long-horizon synthesis, we introduce a Hybrid Context Integration mechanism to maintain subject and scene consistency across minute-level generations. Experiments demonstrate that our method significantly outperforms state-of-the-art methods, offering superior structural control and creative diversity for video synthesis. Our project has been available on: https://martayang.github.io/ONE-SHOT/.

Problem

• compositional video synthesis
• human-environment editing
• spatial decoupling
• generative flexibility
• 3D pre-processing

Innovation

• Spatial-Decoupled Motion Injection
• Dynamic-Grounded-RoPE
• Hybrid Context Integration
• Compositional Video Synthesis
• Video Foundation Models