Network Bending of Diffusion Models for Audio-Visual Generation
This work addresses a creative application for artists by providing a tool for audio-visual generation, but it is incremental as it builds on existing diffusion models and network bending techniques.
The paper tackles the problem of enabling artists to create music visualizations by applying network bending transforms to diffusion models, resulting in continuous, fine-grain control over image generation and the identification of visual effects not easily recreated with standard tools.
In this paper we present the first steps towards the creation of a tool which enables artists to create music visualizations using pre-trained, generative, machine learning models. First, we investigate the application of network bending, the process of applying transforms within the layers of a generative network, to image generation diffusion models by utilizing a range of point-wise, tensor-wise, and morphological operators. We identify a number of visual effects that result from various operators, including some that are not easily recreated with standard image editing tools. We find that this process allows for continuous, fine-grain control of image generation which can be helpful for creative applications. Next, we generate music-reactive videos using Stable Diffusion by passing audio features as parameters to network bending operators. Finally, we comment on certain transforms which radically shift the image and the possibilities of learning more about the latent space of Stable Diffusion based on these transforms.