# US 12,604,032 B2 · Systems and methods for performing padding in coding of a multi-dimensional data set

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,604,032 B2
- Document ID: US12604032B2
- Status: Granted
- Kind: B2
- Application number: 18/617,980
- Filed: 2024-03-27
- Published: 2024-08-08
- Granted: 2026-04-14
- Earliest claimed: 2021-03-26
- Assignees: Sharp Kabushiki Kaisha
- Applicants: Sharp Kabushiki Kaisha
- Inventors: Kiran Mukesh Misra; Tianying Ji; Christopher Andrew Segall
- Web page: https://patentgrep.com/patent/US12604032B2

## CPC

- H04N 19/122: Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type
- H04N 19/563: Motion estimation with padding, i.e. with filling of non-object values in an arbitrarily shaped picture block or region for estimation purposes
- G06T 9/002: {using neural networks}
- H04N 19/11: among a plurality of spatial predictive coding modes
- H04N 19/132: Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking
- H04N 19/176: the region being a block, e.g. a macroblock
- H04N 19/18: the unit being a set of transform coefficients
- H04N 19/59: involving spatial sub-sampling or interpolation, e.g. alteration of picture size or resolution
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method of encoding data includes: receiving an input data set having an arbitrary size about a height dimension and a width dimension; padding the input data set according to a padding function selected from a set of padding functions, such that a data set having a desired size about the height dimension and the width dimension is obtained; generating an output data set by performing a discrete convolution on the obtained data set; and generating a signal providing information corresponding to the output data set and the selected padding function.

## Claims

### Claim 1

A method of encoding video data, the method comprising:
receiving an input data set of video data;
generating an output data set, including output feature maps, by performing a discrete convolution on the input data set, wherein performing a discrete convolution on the input data set includes spatially down-sampling the input data set;
quantizing the output feature maps on a picture-by-picture basis, wherein quantizing the output feature maps on the picture-by-picture basis includes reducing a bit-depth of sample values in the output feature maps to a bit-depth which is the same as the input video data bit-depth by mapping ranges of amplitudes to quantization indices using different quantizers for each picture; and
generating a bitstream by encoding the quantized output feature maps according to a video coding standard utilizing lossy video compression techniques.

### Claim 2

A device comprising one or more processors configured to:
receive an input data set of video data;
generate an output data set, including output feature maps, by performing a discrete convolution on the input data set, wherein performing a discrete convolution on the input data set includes spatially down-sampling the input data set;
quantize the output feature maps on a picture-by-picture basis, wherein quantizing the output feature maps on the picture-by-picture basis includes reducing a bit-depth of sample values in the output feature maps to a bit-depth which is the same as the input video data bit-depth by mapping ranges of amplitudes to quantization indices using different quantizers for each picture; and
generate a bitstream by encoding the quantized output feature maps according to a video coding standard utilizing lossy video compression techniques.

### Claim 3 (depends on 2)

The device of claim 2, wherein the device includes a compression engine.