Back to results
US 12,034,967 B2Granted
Superpixel generation and use
AssigneeNVIDIA Corporation
InventorsVipul Parashar, Vignesh Ungrapalli
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, and techniques to interpolate one or more intermediate images from two or more images is disclosed. In at least one embodiment, a processor includes one or more circuits to interpolate one or more intermediate images from two or more images based, at least in part, on one or more inconsistent flow vectors corresponding to the two or more images.
Claims41
A processor comprising:
- one or more circuits to cause motion of one or more pixels between two or more video frames to be predicted based, at least in part, on a plurality of pixels within a radius of the one or more pixels if one or more forward motion vectors corresponding to the one or more pixels are inconsistent with one or more backward motion vectors corresponding to the one or more pixels.
A system, comprising a memory;
- and
- one or more circuits, the one or more circuits to:
- cause motion of one or more pixels between two or more video frames to be predicted based, at least in part, on a plurality of pixels within a radius of the one or more pixels if one or more forward motion vectors corresponding to the one or more pixels are inconsistent with one or more backward motion vectors corresponding to the one or more pixels.
A method comprising:
- predicting motion of one or more pixels between two or more video frames based, at least in part, on a plurality of pixels within a radius of the one or more pixels if one or more forward motion vectors corresponding to the one or more pixels is inconsistent with one or more backward motion vectors corresponding to the one or more pixels.
A processor comprising one or more circuits to
- train one or more neural networks to perform a machine vision task using a training dataset comprising motion of one or more pixels between two or more video frames that are predicted based at least in part on a plurality of pixels within a radius of the one or more pixels if one or more forward motion vectors corresponding to the one or more pixels are consistent with one or more backward motion vectors corresponding to the one or more pixels.
A processor comprising one or more circuits to:
- use one or more trained neural networks to process an input comprising motion of one or more pixels between two or more video frames that are predicted based at least in part on a plurality of pixels within a radius of the one or more pixels if one or more forward motion vectors corresponding to the one or more pixels are consistent with one or more backward motion vectors corresponding to the one or more pixels.
Dependencies are inferred from the claim text.