Abstract: Apparatuses, systems, and techniques are presented to reconstruct one or more images. In at least one embodiment, one or more neural networks are used to generate one or more images of one or more objects based, at least in part, on input indicating motion of the one or more objects.

Abstract: Apparatuses, systems, and techniques to perform compression of video data using neural networks to facilitate video streaming, such as video conferencing. In at least one embodiment, a sender transmits to a receiver a key frame from video data and one or more keypoints identified by a neural network from said video data, and a receiver reconstructs video data using said key frame and one or more received keypoints.

Abstract: An apparatus, computer readable medium, and method are disclosed for decompressing compressed geometric data stored in a lossless compression format. The compressed geometric data resides within a compression block sized according to a system cache line. An indirection technique maps a global identifier value in a linear identifier space to corresponding variable rate compressed data. The apparatus may include decompression circuitry within a graphics processing unit configured to perform ray-tracing.

Abstract: A few-shot, unsupervised image-to-image translation (“FUNIT”) algorithm is disclosed that accepts as input images of previously-unseen target classes. These target classes are specified at inference time by only a few images, such as a single image or a pair of images, of an object of the target type. A FUNIT network can be trained using a data set containing images of many different object classes, in order to translate images from one class to another class by leveraging few input images of the target class. By learning to extract appearance patterns from the few input images for the translation task, the network learns a generalizable appearance pattern extractor that can be applied to images of unseen classes at translation time for a few-shot image-to-image translation task.

Abstract: An apparatus, computer readable medium, and method are disclosed for decompressing compressed geometric data stored in a lossless compression format. The compressed geometric data resides within a compression block sized according to a system cache line. An indirection technique maps a global identifier value in a linear identifier space to corresponding variable rate compressed data. The apparatus may include decompression circuitry within a graphics processing unit configured to perform ray-tracing.

Abstract: A method of generating an image. The method includes simulating a presence of at least one light source within a virtualized three dimensional space. Within the virtualized three dimensional space, a light sensing plane is defined. The light sensing plane includes a matrix of a number of pixels to be displayed on a display screen. The method further includes using a light transport procedure, computing a gradient value for each pixel of the matrix to produce a number of gradient values. The gradient computation involves selecting a plurality of light path pairs that contribute to a pixel wherein the selection is biased towards selection of more light paths that pass through pixels having larger gradient values. The plurality of gradient values are converted to a plurality of light intensity values which represent the image.

Abstract: A method for reducing the number of samples tested for rendering a screen space region of an image includes constructing a trajectory of a primitive extending in an image which is to be rendered. A bounding volume is constructed for a screen space region of the image, the bounding volume characterized as having a bound in a non-screen space dimension which is defined as a function of the primitive's trajectory. The bounding volume is further characterized as overlapping a portion of the screen space region which is to be rendered. One or more sample points which are located within the screen space region, and which are not overlapped by the bounding volume are excluded from testing.

Abstract: A method for reducing the number of samples tested for rendering a screen space region of an image includes constructing a trajectory of a primitive in a three dimensional coordinate system, the coordinate system including a screen space dimension, a lens dimension and a time dimension. A bounding volume is constructed for a screen space region which is to be rendered, the bounding volume overlapping a portion of the screen space region. The bounding volume is defined according to a plurality of bounding planes which extend in the three dimensional coordinate system, whereby the bounding planes are determined as a function of the trajectory of the primitive. One or more sample points which are located within the screen space region, and which are not overlapped by the bounding volume are excluded from testing.

Abstract: A method for reducing the number of samples tested for rendering a screen space region of an image includes constructing a trajectory of a primitive extending within an image which is to be rendered. A bounding volume is constructed for a screen space region of the image, the bounding volume characterized as having a bound in a non-screen space dimension which is defined as a function of the primitive's trajectory. The bounding volume is further characterized as overlapping a portion of the screen space region which is to be rendered. One or more sample points which are located within the screen space region, and which are not overlapped by the bounding volume are excluded from testing.

Abstract: A bounding box-based method for reducing the number of samples tested for rendering a screen space region of an image includes determining a trajectory of a primitive in screen space for an image which is to be rendered and constructing an axis-aligned bounding box for the screen space region. The axis-aligned bounding box includes a bound in a non-screen dimension that is defined as a function of the screen space trajectory of the primitive, and overlaps a portion of the screen space region. One or more sample points which are located within the screen space region, and which are not overlapped by the axis-aligned bounding box are excluded from testing.

Abstract: A method of generating an image. The method includes simulating a presence of at least one light source within a virtualized three dimensional space. Within the virtualized three dimensional space, a light sensing plane is defined. The light sensing plane includes a matrix of a number of pixels to be displayed on a display screen. The method further includes using a light transport procedure, computing a gradient value for each pixel of the matrix to produce a number of gradient values. The gradient computation involves selecting a plurality of light path pairs that contribute to a pixel wherein the selection is biased towards selection of more light paths that pass through pixels having larger gradient values. The plurality of gradient values are converted to a plurality of light intensity values which represent the image.

Abstract: A method for reducing the number of samples tested for rendering a screen space region of an image includes constructing a trajectory of a primitive extending in an image which is to be rendered. A bounding volume is constructed for a screen space region of the image, the bounding volume characterized as having a bound in a non-screen space dimension which is defined as a function of the primitive's trajectory. The bounding volume is further characterized as overlapping a portion of the screen space region which is to be rendered. One or more sample points which are located within the screen space region, and which are not overlapped by the bounding volume are excluded from testing.