论文标题
通过组成神经辐射场学习多对象动力学
Learning Multi-Object Dynamics with Compositional Neural Radiance Fields
论文作者
论文摘要
我们提出了一种从基于隐式对象编码器，神经辐射场（NERFS）和图形神经网络的图像观测值中学习组成多对象动力学模型的方法。由于其强劲的3D先验，NERF已成为代表场景的流行选择。但是，大多数NERF方法都在一个场景上进行了训练，以全球模型代表整个场景，从而对新型场景进行概括，其中包含不同数量的对象，具有挑战性。取而代之的是，我们提出一个以对象为中心的自动编码器框架，该框架将场景的多个视图映射到分别表示每个对象的一组潜在向量。潜在矢量可以将单个nerf参数化，可以从中重建场景。根据那些潜在向量，我们在潜在空间中训练图形神经网络动力学模型，以实现动力学预测的组成性。我们方法的一个关键特征是，潜在向量被迫通过NERF解码器编码3D信息，这使我们能够在学习动力学模型中纳入结构先验，从而使长期预测与多个基线相比更加稳定。模拟和现实世界的实验表明，我们的方法可以建模和学习构图场景的动态，包括刚性和可变形对象。视频：https：//dannydriess.github.io/compnerfdyn/
We present a method to learn compositional multi-object dynamics models from image observations based on implicit object encoders, Neural Radiance Fields (NeRFs), and graph neural networks. NeRFs have become a popular choice for representing scenes due to their strong 3D prior. However, most NeRF approaches are trained on a single scene, representing the whole scene with a global model, making generalization to novel scenes, containing different numbers of objects, challenging. Instead, we present a compositional, object-centric auto-encoder framework that maps multiple views of the scene to a set of latent vectors representing each object separately. The latent vectors parameterize individual NeRFs from which the scene can be reconstructed. Based on those latent vectors, we train a graph neural network dynamics model in the latent space to achieve compositionality for dynamics prediction. A key feature of our approach is that the latent vectors are forced to encode 3D information through the NeRF decoder, which enables us to incorporate structural priors in learning the dynamics models, making long-term predictions more stable compared to several baselines. Simulated and real world experiments show that our method can model and learn the dynamics of compositional scenes including rigid and deformable objects. Video: https://dannydriess.github.io/compnerfdyn/
