Amazon Technologies, Inc.

Sergey Prokudin

Javier Romero Gonzalez-Nicolas

Michael Julian Black

Christoph Lassner

Systems and methods are provided for generating an image of a posed human figure or other subject using a neural network that is trained to translate a set of points to realistic images by reconstructing projected surfaces directly in the pixel space or image space. Input to the image generation process may include parameterized control features, such as body shape parameters, pose parameters and/or a virtual camera position. These input parameters may be applied to a three-dimensional model that is used to generate the set of points, such as a sparsely populated image of color and depth information at vertices of the three-dimensional model, before additional image generation occurs directly in the image space. The visual appearance or identity of the synthesized human in successive output images may remain consistent, such that the output is both controllable and predictable.

Image generation from 3D model using neural network

A system configured to process an input point cloud, which represents an object using unstructured data points, to generate a feature vector that has an ordered structure and a fixed length. The system may process the input point cloud using a basis point set to generate the feature vector. For example, for each basis point in the basis point set, the system may identify a closest data point in the point cloud data and store a distance value or other information associated with the closest data point in the feature vector. The system may process the feature vector using a trained model to generate output data, such as performing point cloud registration to generate mesh data, point cloud classification to generate classification data, and/or the like.

Growing community of inventors

Tubingen, Germany

Sergey Prokudin