English

Efficient Training for Human Video Generation with Entropy-Guided Prioritized Progressive Learning

Computer Vision and Pattern Recognition 2025-11-27 v1 Artificial Intelligence

Abstract

Human video generation has advanced rapidly with the development of diffusion models, but the high computational cost and substantial memory consumption associated with training these models on high-resolution, multi-frame data pose significant challenges. In this paper, we propose Entropy-Guided Prioritized Progressive Learning (Ent-Prog), an efficient training framework tailored for diffusion models on human video generation. First, we introduce Conditional Entropy Inflation (CEI) to assess the importance of different model components on the target conditional generation task, enabling prioritized training of the most critical components. Second, we introduce an adaptive progressive schedule that adaptively increases computational complexity during training by measuring the convergence efficiency. Ent-Prog reduces both training time and GPU memory consumption while maintaining model performance. Extensive experiments across three datasets, demonstrate the effectiveness of Ent-Prog, achieving up to 2.2×\times training speedup and 2.4×\times GPU memory reduction without compromising generative performance.

Keywords

Cite

@article{arxiv.2511.21136,
  title  = {Efficient Training for Human Video Generation with Entropy-Guided Prioritized Progressive Learning},
  author = {Changlin Li and Jiawei Zhang and Shuhao Liu and Sihao Lin and Zeyi Shi and Zhihui Li and Xiaojun Chang},
  journal= {arXiv preprint arXiv:2511.21136},
  year   = {2025}
}

Comments

Project page: https://github.com/changlin31/Ent-Prog

R2 v1 2026-07-01T07:55:44.366Z