In this paper, we introduce Matten, a cutting-edge latent diffusion model with Mamba-Attention architecture for video generation. With minimal computational cost, Matten employs spatial-temporal attention for local video content modeling and bidirectional Mamba for global video content modeling. Our comprehensive experimental evaluation demonstrates that Matten has competitive performance with the current Transformer-based and GAN-based models in benchmark performance, achieving superior FVD scores and efficiency. Additionally, we observe a direct positive correlation between the complexity of our designed model and the improvement in video quality, indicating the excellent scalability of Matten.
@article{arxiv.2405.03025,
title = {Matten: Video Generation with Mamba-Attention},
author = {Yu Gao and Jiancheng Huang and Xiaopeng Sun and Zequn Jie and Yujie Zhong and Lin Ma},
journal= {arXiv preprint arXiv:2405.03025},
year = {2024}
}