论文标题
相关：使用结构性潜在空间，物理上合理的多对象场景合成
RELATE: Physically Plausible Multi-Object Scene Synthesis Using Structured Latent Spaces
论文作者
论文摘要
我们提出了联系，该模型学会生成多个交互对象的物理上合理的场景和视频。与其他生成方法类似，RELATE是在原始的，未标记的数据上端到端训练的。将以对象为中心的GAN公式与一个明确说明单个对象之间相关性的模型相结合。这使该模型可以从物理上解剖的参数化中生成逼真的场景和视频。此外，我们表明对对象相关进行建模是为了学习解除对象位置和身份的必要条件。我们发现，与物理上现实的场景编辑相关，并且在合成（CLEVR，Shapestacks）和现实世界中的数据（CARS）中，它在以对象为中心的场景中的先前艺术（CARS）都大大优于先前的艺术。此外，与以对象为中心的生成建模中的最新方法相反，它还与动态场景自然扩展并生成高视觉保真度的视频。源代码，数据集和更多结果可从http://geometry.cs.ucl.ac.uk/projects/2020/relate/获得。
We present RELATE, a model that learns to generate physically plausible scenes and videos of multiple interacting objects. Similar to other generative approaches, RELATE is trained end-to-end on raw, unlabeled data. RELATE combines an object-centric GAN formulation with a model that explicitly accounts for correlations between individual objects. This allows the model to generate realistic scenes and videos from a physically-interpretable parameterization. Furthermore, we show that modeling the object correlation is necessary to learn to disentangle object positions and identity. We find that RELATE is also amenable to physically realistic scene editing and that it significantly outperforms prior art in object-centric scene generation in both synthetic (CLEVR, ShapeStacks) and real-world data (cars). In addition, in contrast to state-of-the-art methods in object-centric generative modeling, RELATE also extends naturally to dynamic scenes and generates videos of high visual fidelity. Source code, datasets and more results are available at http://geometry.cs.ucl.ac.uk/projects/2020/relate/.
