Tuning-Free Noise Rectification

for High Fidelity Image-to-Video Generation

Weijie Li,
Litong Gong,
Yiran Zhu,
Fanda Fan,
Biao Wang,
Tiezheng Ge,
Bo Zheng
Alibaba Inc.
arXiv Code YouTube

Comparison

Noise Rectification is a simple but effective method for image-to-video generation in open domains, and is tuning-free and plug-and-play. Below are several comparisons between our method and other methods.

Image-to-Video Examples

Abstract

Image-to-video (I2V) generation tasks always suffer from keeping high fidelity in the open domains. Traditional image animation techniques primarily focus on specific domains such as faces or human poses, making them difficult to generalize to open domains. Several recent I2V frameworks based on diffusion models can generate dynamic content for open domain images but fail to maintain fidelity. We found that two main factors of low fidelity are the loss of image details and the noise prediction biases during the denoising process. To this end, we propose an effective method that can be applied to mainstream video diffusion models. This method achieves high fidelity based on supplementing more precise image information and noise rectification. Specifically, given a specified image, our method first adds noise to the input image latent to keep more details, then denoises the noisy latent with proper rectification to alleviate the noise prediction biases. Our method is tuning-free and plug-and-play. The experimental results demonstrate the effectiveness of our approach in improving the fidelity of generated videos.

Framework


The framework of our tuning-free image-to-video method. It represents the inference pipeline, where the input image is noised into the initial latent and the predicted noise of the inflated 3D U-Net will be rectified during the denoising process.

BibTeX

@misc{noise2024,
      title={Tuning-Free Noise Rectification for High Fidelity Image-to-Video Generation},
      author={Li, Weijie and Gong, Litong and Zhu, Yiran and  Fan, Fanda and Wang, Biao and Ge, Tiezheng and Zheng, Bo},
      year={2024},
      eprint={arXiv:2403.02827},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}