Back to results
US 12,657,788 B2Granted
Systems and methods for recognising hand-drawn shapes
AssigneeCanva Pty Ltd
InventorsKevin Andrew Wu Won, Kerry Jayne Halupka, Rowan James Katekar
Life of this patent
Estimating the term…
Abstract
Described herein is a computer implemented method. The method includes receiving, via an input device, first user input drawing an input shape and generating, based on the first user input, original drawing data that includes an ordered set of points that define the input shape. The original drawing is processed to generate an input vector which also includes an ordered set of points. The input shape is then classified as a first template shape by processing the input vector using a machine learning model. A new shape is then generated based on the first template shape and the original drawing data.
Claims19
A computer implemented method including:
- receiving, via an input device, first user input drawing an input shape;
- sampling the first user input at a sampling rate to generate original drawing data, wherein the original drawing data includes an ordered set of points and wherein each point in the original drawing data corresponds to a position of the first user input at a particular time;
- processing the original drawing data to generate an input vector, wherein the input vector includes an ordered set of points and each point in the ordered set of points of the input vector is a coordinate pair;
- classifying the input shape into a first class that corresponds to a first template shape by processing the input vector using a machine learning model, wherein the machine learning model is a neural network type machine learning model;
- generating a new shape based on the first template shape;
- and
- outputting the new shape, and wherein:
- processing the original drawing data to generate the input vector includes processing the original drawing data to generate first standardised drawing data:
- the first standardised drawing data includes an ordered set of points and has a first defined number of points;
- and
- processing the original drawing data to generate the first standardised drawing data includes processing the original drawing data according to a line standardisation method that includes:
- performing one or more top level iterations, each top level iteration including processing the original drawing data to determine a maximally distal point of the original drawing data;
- and
- adding the maximally distal point determined at each top level iteration to a result set;
- and
- on completion of the one or more top level iterations, generating the first standardised drawing data based on the result set.
13 more elements
A computer processing system including:
- a processing unit;
- and
- a non-transitory computer-readable storage medium storing instructions, which when executed by the processing unit, cause the processing unit to perform a method including:
- receiving first user input drawing an input shape;
- sampling the first user input at a sampling rate to generate original drawing data, wherein the original drawing data includes an ordered set of points and wherein each point in the original drawing data corresponds to a position of the first user input at a particular time;
- processing the original drawing data to generate an input vector, wherein the input vector includes an ordered set of points and each point in the ordered set of points of the input vector is a coordinate pair;
- classifying the input shape into a first class that corresponds to a first template shape by processing the input vector using a machine learning model, wherein the machine learning model is a neural network type machine learning model;
- generating a new shape based on the first template shape;
- and
- outputting the new shape, and wherein:
- processing the original drawing data to generate the input vector includes processing the original drawing data to generate first standardised drawing data;
- the first standardised drawing data includes an ordered set of points and has a first defined number of points;
- and
- processing the original drawing data to generate the first standardised drawing data includes processing the original drawing data according to a line standardisation method that includes:
- performing one or more top level iterations, each top level iteration including processing the original drawing data to determine a maximally distal point of the original drawing data;
- and
- adding the maximally distal point determined at each top level iteration to a result set;
- and
- on completion of the one or more top level iterations, generating the first standardised drawing data based on the result set.
16 more elements
Non-transitory storage medium storing instructions executable by processing unit to cause the processing unit to perform a method including:
- receiving first user input drawing an input shape;
- sampling the first user input at a sampling rate to generate original drawing data, wherein the original drawing data includes an ordered set of points and wherein each point in the original drawing data corresponds to a position of the first user input at a particular time;
- processing the original drawing data to generate an input vector, wherein the input vector includes an ordered set of points and each point in the ordered set of points of the input vector is a coordinate pair;
- classifying the input shape into a first class that corresponds to a first template shape by processing the input vector using a machine learning model, wherein the machine learning model is a neural network type machine learning model;
- generating a new shape based on the first template shape;
- and
- outputting the new shape, and wherein:
- processing the original drawing data to generate the input vector includes processing the original drawing data to generate first standardised drawing data;
- the first standardised drawing data includes an ordered set of points and has a first defined number of points;
- and
- processing the original drawing data to generate the first standardised drawing data includes processing the original drawing data according to a line standardisation method that includes:
- performing one or more top level iterations, each top level iteration including processing the original drawing data to determine a maximally distal point of the original drawing data;
- and
- adding the maximally distal point determined at each top level iteration to a result set;
- and
- on completion of the one or more top level iterations, generating the first standardised drawing data based on the result set.
13 more elements
Dependencies are inferred from the claim text.