Title: Controllable Video Generation by Learning the Underlying Dynamical System with Neural ODE
Abstract: Videos depict the change of complex dynamical systems over time in the form of discrete image sequences. Generating controllable videos by learning the dynamical system is an important yet underexplored topic in the computer vision community. In this talk, I will present a novel framework, TiV-ODE, for generating highly controllable videos from a static image and a text caption. Specifically, our framework leverages the ability of Neural Ordinary Differential Equations (Neural ODEs) to represent complex dynamical systems as a set of nonlinear ordinary differential equations. The resulting framework is capable of generating videos with both desired dynamics and content. Experiments demonstrate the ability of the proposed method in generating highly controllable and visually consistent videos, and its capability of modeling dynamical systems.
Title: Mass-Spring-Charge Simulation
Abstract: In this talk, I will present a physics simulation of the Coulomb potential using an implicit integration method in a mass-spring system. The implicit integration method allows for much larger time steps while maintaining numerical stability, avoiding the issue of energy explosions associated with explicit methods like Verlet. We will discuss the accuracy and efficiency of our approach, highlighting the trade-offs between numerical stability and accuracy. Our simulation has potential applications in various fields, such as molecular dynamics and material science, where electrostatic simulations play a significant role in understanding physical systems.