Back to results
US 12,749,146 B2Granted
Generating image blending weights
AssigneeNVIDIA Corporation
InventorPekka Janis
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, and techniques are presented to reconstruct one or more images. In at least one embodiment, one or more anisotropic filters are used to determine one or more pixel blending weights.
Claims33
One or more processors, comprising circuitry to:
- cause one or more neural networks to generate one or more anisotropic filters based, at least in part, on one or more video frames and one or more upsampled video frames;
- use the one or more neural networks to generate one or more pixel blending weights based, at least in part, on the one or more anisotropic filters;
- and
- blend two or more pixels of the one or more video frames and the one or more upsampled video frames using the one or more pixel blending weights.
1 more element
A system comprising one or more processors to:
- cause one or more neural networks to generate one or more anisotropic filters based, at least in part, on one or more video frames and one or more upsampled video frames;
- use the one or more neural networks to generate one or more pixel blending weights based, at least in part, on the one or more anisotropic filters;
- and
- blend two or more pixels of the one or more video frames and the one or more upsampled video frames using the one or more pixel blending weights.
1 more element
A method comprising:
- causing one or more neural networks to generate one or more anisotropic filters based, at least in part, on one or more video frames and one or more upsampled video frames;
- using the one or more neural networks to generate one or more pixel blending weights based, at least in part, on the one or more anisotropic filters;
- and
- blending two or more pixels of the one or more video frames and the one or more upsampled video frames using the one or more pixel blending weights.
1 more element
A non-transitory machine-readable medium having stored thereon a set of instructions, which if performed by one or more processors, cause the one or more processors to at least:
- cause one or more neural networks to generate one or more anisotropic filters based, at least in part, on one or more video frames and one or more upsampled video frames;
- use the one or more neural networks to generate one or more pixel blending weights based, at least in part, on the one or more anisotropic filters;
- and
- blend two or more pixels of the one or more video frames and the one or more upsampled video frames using the one or more pixel blending weights.
1 more element
An image reconstruction system, comprising:
- one or more processors to:
- cause one or more neural networks to generate one or more anisotropic filters based, at least in part, on one or more video frames and one or more upsampled video frames;
- use the one or more neural networks to generate one or more pixel blending weights based, at least in part, on the one or more anisotropic filters;
- and
- blend two or more pixels of the one or more video frames and the one or more upsampled video frames using the one or more pixel blending weights;
- and
- memory for storing the one or more anisotropic filters.
4 more elements
Dependencies are inferred from the claim text.