ReImagine: Rethinking Controllable High-Quality Human Video Generation via Image-First Synthesis

arXiv:2604.19720v1 Announce Type: new Abstract: Human video generation remains challenging due to the difficulty of jointly modeling human appearance, motion, and camera viewpoint under limited multi-view data. Existing methods often address these factors separately, resulting in limited controllability or reduced visual quality. We revisit this problem from an image-first perspective, where high-quality human appearance is learned via image generation and used as a prior for video synthesis, decoupling appearance modeling from temporal consistency. We propose a pose- and viewpoint-controllable pipeline that combines a pretrained image backbone with SMPL-X-based motion guidance, together with a training-free temporal refinement stage based on a pretrained video diffusion model. Our method produces high-quality, temporally consistent videos under diverse poses and viewpoints. We also release a canonical human dataset and an auxiliary model for compositional human image synthesis. Code and data are publicly available at https://github.com/Taited/ReImagine.

Related

• TS-Attn: Temporal-wise Separable Attention for Multi-Event Video Generation
• PoseGen: In-Context LoRA Finetuning for Pose-Controllable Long Human Video Generation
• GenHSI: Controllable Generation of Human-Scene Interaction Videos
• Prompt-to-Gesture: Measuring the Capabilities of Image-to-Video Deictic Gesture Generation

Source: arXiv cs.CV | 2026-04-22