Gordon Wetzstein: Neural Scene Representation and Rendering скачать в хорошем качестве

Gordon Wetzstein: Neural Scene Representation and Rendering

In this talk, we discuss recent progress on implicitly defined neural-network-parameterized signal representation, processing, and rendering techniques. Such representation networks model 3D scenes in a continuous and differentiable manner, for example using signed distance functions or neural volumes. Differentiable renderers, including sphere tracing and neural volume rendering (i.e., NeRF), allow for these representations to be learned from partial 2D observations. In this context we will discuss recent advances in efficient representation network architectures, rendering algorithms, generalization strategies, and generative adversarial approaches for applications ranging from novel view synthesis to unconditional 3D scene generation.