Motion Modes: What Could Happen Next?
This addresses the challenge of generating diverse object motions in complex scenes for applications in video generation and animation, representing a novel method rather than an incremental improvement.
The paper tackles the problem of predicting diverse object motions from a single static image by introducing Motion Modes, a training-free approach that uses a pre-trained image-to-video generator to discover plausible motions, resulting in realistic and varied animations that surpass previous methods and human predictions in plausibility and diversity.
Predicting diverse object motions from a single static image remains challenging, as current video generation models often entangle object movement with camera motion and other scene changes. While recent methods can predict specific motions from motion arrow input, they rely on synthetic data and predefined motions, limiting their application to complex scenes. We introduce Motion Modes, a training-free approach that explores a pre-trained image-to-video generator's latent distribution to discover various distinct and plausible motions focused on selected objects in static images. We achieve this by employing a flow generator guided by energy functions designed to disentangle object and camera motion. Additionally, we use an energy inspired by particle guidance to diversify the generated motions, without requiring explicit training data. Experimental results demonstrate that Motion Modes generates realistic and varied object animations, surpassing previous methods and even human predictions regarding plausibility and diversity. Project Webpage: https://motionmodes.github.io/