Neural Implicit Representations for Physical Parameter Inference from a Single Video

@article{Hofherr2022NeuralIR,
  title={Neural Implicit Representations for Physical Parameter Inference from a Single Video},
  author={Florian Hofherr and Lukas Koestler and Florian Bernard and Daniel Cremers},
  journal={ArXiv},
  year={2022},
  volume={abs/2204.14030}
}
Neural networks have recently been used to analyze diverse physical systems and to identify the underlying dynamics. While existing methods achieve impressive results, they are limited by their strong demand for training data and their weak generalization abilities to out-of-distribution data. To overcome these limitations, we propose to combine neural implicit representations for appearance modeling with neural ordinary differential equations (ODEs) for modelling planar physical phenomena to…