Receptive-field-conforming convolutional models for video coding
US11310498B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Nov 2, 2020 |
| Grant date | Apr 19, 2022 |
| Priority date | — |
| Expiry date | Nov 2, 2040 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04N19/19
- WIPO fieldAudio-visual technology
- WIPO sectorElectrical engineering
Abstract
An apparatus for encoding a block of a picture includes a convolutional neural network (CNN) for determining a block partitioning of the block, the block having an N×N size and a smallest partition determined by the CNN being of size S×S. The CNN includes feature extraction layers; a concatenation layer that receives, from the feature extraction layers, first feature maps of the block, where each first feature map of the first feature maps is of the smallest possible partition size S×S of the block; and at least one classifier that is configured to infer partition decisions for sub-blocks of size (αS)×(αS) of the block, where α is a power of 2.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.