Loopy: Taming Audio-Driven Portrait Avatar with Long-Term Motion Dependency
This work addresses the challenge of achieving natural motion in audio-driven portrait generation for applications like virtual avatars, though it appears incremental by building on diffusion-based techniques.
The paper tackles the problem of generating natural human motion in audio-driven portrait avatars by proposing Loopy, an end-to-end audio-only conditioned video diffusion model that leverages long-term motion dependencies, resulting in more lifelike and high-quality outputs compared to existing methods.
With the introduction of diffusion-based video generation techniques, audio-conditioned human video generation has recently achieved significant breakthroughs in both the naturalness of motion and the synthesis of portrait details. Due to the limited control of audio signals in driving human motion, existing methods often add auxiliary spatial signals to stabilize movements, which may compromise the naturalness and freedom of motion. In this paper, we propose an end-to-end audio-only conditioned video diffusion model named Loopy. Specifically, we designed an inter- and intra-clip temporal module and an audio-to-latents module, enabling the model to leverage long-term motion information from the data to learn natural motion patterns and improving audio-portrait movement correlation. This method removes the need for manually specified spatial motion templates used in existing methods to constrain motion during inference. Extensive experiments show that Loopy outperforms recent audio-driven portrait diffusion models, delivering more lifelike and high-quality results across various scenarios.