Abstract: Performance of a neural network is usually a function of the capacity, or complexity, of the neural network, including the depth of the neural network (i.e. the number of layers in the neural network) and/or the width of the neural network (i.e. the number of hidden channels). However, improving performance of a neural network by simply increasing its capacity has drawbacks, the most notable being the increased computational cost of a higher-capacity neural network. Since modern neural networks are configured such that the same neural network is evaluated regardless of the input, a higher capacity neural network means a higher computational cost incurred per input processed. The present disclosure provides for a multi-layer neural network that allows for dynamic path selection through the neural network when processing an input, which in turn can allow for increased neural network capacity without incurring the typical increased computation cost associated therewith.

Abstract: Synthesis of high-quality 3D shapes with smooth surfaces has various creative and practical use cases, such as 3D content creation and CAD modeling. A vector field decoder neural network is trained to predict a generative vector field (GVF) representation of a 3D shape from a latent representation (latent code or feature volume) of the 3D shape. The GVF representation is agnostic to surface orientation, all dimensions of the vector field vary smoothly, the GVF can represent both watertight and non-watertight 3D shapes, and there is a one-to-one mapping between a predicted 3D shape and the ground truth 3D shape (i.e., the mapping is bijective). The vector field decoder can synthesize 3D shapes in multiple categories and can also synthesize 3D shapes for objects that were not included in the training dataset. In other words, the vector field decoder is also capable of zero-shot generation.

Abstract: The technology described herein is directed to a cross-domain training framework that iteratively trains a domain adaptive refinement agent to refine low quality real-world image acquisition data, e.g., depth maps, when accompanied by corresponding conditional data from other modalities, such as the underlying images or video from which the image acquisition data is computed. The cross-domain training framework includes a shared cross-domain encoder and two conditional decoder branch networks, e.g., a synthetic conditional depth prediction branch network and a real conditional depth prediction branch network. The shared cross-domain encoder converts synthetic and real-world image acquisition data into synthetic and real compact feature representations, respectively.

Abstract: Apparatuses, systems, and techniques are presented to generate images. In at least one embodiment, one or more neural networks are used to generate one or more images based, at least in part, upon one or more semantic features projected from a three-dimensional environment.

Abstract: The technology described herein is directed to a cross-domain training framework that iteratively trains a domain adaptive refinement agent to refine low quality real-world image acquisition data, e.g., depth maps, when accompanied by corresponding conditional data from other modalities, such as the underlying images or video from which the image acquisition data is computed. The cross-domain training framework includes a shared cross-domain encoder and two conditional decoder branch networks, e.g., a synthetic conditional depth prediction branch network and a real conditional depth prediction branch network. The shared cross-domain encoder converts synthetic and real-world image acquisition data into synthetic and real compact feature representations, respectively.