Neural Re-Simulation for Generating Bounces in Single Images

Abstract

We introduce a method to generate videos of dynamic virtual objects plausibly interacting via collisions with a still image's environment. Given a starting trajectory, physically simulated with the estimated geometry of a single, static input image, we learn to 'correct' this trajectory to a visually plausible one via a neural network. The neural network can then be seen as learning to 'correct' traditional simulation output, generated with incomplete and imprecise world information, to obtain context-specific, visually plausible re-simulated output - a process we call neural re-simulation. We train our system on a set of 50k synthetic scenes where a virtual moving object (ball) has been physically simulated. We demonstrate our approach on both our synthetic dataset and a collection of real-life images depicting everyday scenes, obtaining consistent improvement over baseline alternatives throughout.

Cite

Text

Innamorati et al. "Neural Re-Simulation for Generating Bounces in Single Images." Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019. doi:10.1109/ICCV.2019.00881

Markdown

[Innamorati et al. "Neural Re-Simulation for Generating Bounces in Single Images." Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019.](https://mlanthology.org/iccv/2019/innamorati2019iccv-neural/) doi:10.1109/ICCV.2019.00881

BibTeX

@inproceedings{innamorati2019iccv-neural,
  title     = {{Neural Re-Simulation for Generating Bounces in Single Images}},
  author    = {Innamorati, Carlo and Russell, Bryan and Kaufman, Danny M. and Mitra, Niloy J.},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision},
  year      = {2019},
  doi       = {10.1109/ICCV.2019.00881},
  url       = {https://mlanthology.org/iccv/2019/innamorati2019iccv-neural/}
}