arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1908.06217cs.CV
收藏

神经网络再模拟在单幅图像中产生反弹

Neural Re-Simulation for Generating Bounces in Single Images

Adobe Research(奥多比研究院);University College London(伦敦大学学院)

Carlo Innamorati, Bryan Russell, Danny M. Kaufman, and Niloy J. Mitra

摘要:

我们介绍了一种方法来生成视频的动态虚拟对象通过碰撞与静态图像的环境,可扩展地相互作用。给定一个起始轨迹,用单个静态输入图像的估计几何形状进行物理模拟,我们通过神经网络学习将该轨迹“校正”为视觉上合理的轨迹。然后,神经网络可以被看作是学习“校正”传统的模拟输出,用不完整和不精确的世界信息生成,以获得上下文特定的,视觉上合理的重新模拟输出,我们称之为神经重新模拟的过程。我们训练我们的系统上的一组50k的合成场景,其中一个虚拟的移动物体(球)已被物理模拟。我们在我们的合成数据集和描绘日常场景的真实图像集合上展示了我们的方法,在整个基线替代方案上获得了一致的改进。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑