MambaTalk: Efficient Holistic Gesture Synthesis with Selective State Space Models
This work addresses efficiency and quality issues in gesture synthesis for applications like film, robotics, and virtual reality, representing an incremental improvement over existing methods.
The paper tackles the challenge of generating long and diverse gesture sequences with low latency in gesture synthesis by introducing MambaTalk, a method based on state space models with a two-stage modeling strategy and discrete motion priors. The result shows that it matches or exceeds state-of-the-art performance in experiments.
Gesture synthesis is a vital realm of human-computer interaction, with wide-ranging applications across various fields like film, robotics, and virtual reality. Recent advancements have utilized the diffusion model and attention mechanisms to improve gesture synthesis. However, due to the high computational complexity of these techniques, generating long and diverse sequences with low latency remains a challenge. We explore the potential of state space models (SSMs) to address the challenge, implementing a two-stage modeling strategy with discrete motion priors to enhance the quality of gestures. Leveraging the foundational Mamba block, we introduce MambaTalk, enhancing gesture diversity and rhythm through multimodal integration. Extensive experiments demonstrate that our method matches or exceeds the performance of state-of-the-art models. Our project is publicly available at https://kkakkkka.github.io/MambaTalk