Latent Neural Differential Equations for Video Generation
This work addresses the challenge of temporal modeling in video generation for AI and multimedia applications, representing an incremental improvement by replacing older temporal generators with a simpler method.
The paper tackles video generation by using Neural Differential Equations to model temporal dynamics, achieving a new state-of-the-art Inception Score of 15.20 for 64x64 pixel unconditional video generation while reducing parameters and maintaining similar run times.
Generative Adversarial Networks have recently shown promise for video generation, building off of the success of image generation while also addressing a new challenge: time. Although time was analyzed in some early work, the literature has not adequately grown with temporal modeling developments. We study the effects of Neural Differential Equations to model the temporal dynamics of video generation. The paradigm of Neural Differential Equations presents many theoretical strengths including the first continuous representation of time within video generation. In order to address the effects of Neural Differential Equations, we investigate how changes in temporal models affect generated video quality. Our results give support to the usage of Neural Differential Equations as a simple replacement for older temporal generators. While keeping run times similar and decreasing parameter count, we produce a new state-of-the-art model in 64$\times$64 pixel unconditional video generation, with an Inception Score of 15.20.