# US 2021/0232774 A1 · Mechanism to facilitate image translation

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

- Number: US 2021/0232774 A1
- Document ID: US20210232774A1
- Status: Published application
- Kind: A1
- Application number: 16/774,341
- Filed: 2020-01-28
- Published: 2021-07-29
- Earliest claimed: 2020-01-28
- Assignees: salesforce.com, inc.
- Applicants: salesforce.com, inc.
- Inventors: Amit Gupta
- Web page: https://patentgrep.com/patent/US20210232774A1

## CPC

- G06F 40/47: Machine-assisted translation, e.g. using translation memory
- G06F 40/58: Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
- G06N 20/00: Machine learning
- G06T 7/136: involving thresholding
- G06V 30/2445: {Alphabet recognition, e.g. Latin, Kanji or Katakana}
- G06V 30/246: using linguistic properties, e.g. specific for English or German language
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Techniques and structures to facilitate conversion of a workflow process is disclosed. The techniques include receiving an image, identifying one or more objects included in the image, identifying one or more properties associated with each of the one or more objects, generating a matrix including data including the identified objects and associated properties and processing the matrix at a machine learning model to determine whether the image is to be translated based on a determination that one or more objects and associated properties within the image are required to be translated.

## Claims

### Claim 1

A method to facilitate translation of an image, comprising:
receiving an image;
identifying one or more objects included in the image;
identifying one or more properties associated with each of the one or more objects;
generating a matrix including data including the identified objects and associated properties; and
processing the matrix at a machine learning model to determine whether the image is to be translated based on a determination that one or more objects and associated properties within the image are required to be translated.

### Claim 2 (depends on 1)

The method of claim 1, wherein processing the image at the machine learning model comprises generating a property weight associated with each property.

### Claim 3 (depends on 2)

The method of claim 2, wherein processing the image at the machine learning model further comprises:
determining, based on the property weights, one or more objects in the image that require translation; and
generating a list of properties associated with each of the one or more objects that require translation.

### Claim 4 (depends on 3)

The method of claim 3, wherein an object requires translation upon a determination that an average of property weights of the object is greater than an average weight threshold value or an individual property weight is greater than an individual weight threshold value.

### Claim 5 (depends on 4)

The method of claim 4, wherein processing the image at the machine learning model further comprises determining, based on the property weights associated with the list of properties, whether the image is to be translated.

### Claim 6 (depends on 5)

The method of claim 5, wherein an image requires translation upon a determination that an average of property weights in the list of properties is greater than an average weight threshold value or an individual property weight is greater than an individual weight threshold value.

### Claim 7 (depends on 2)

The method of claim 2, further comprising training the machine learning model.

### Claim 8 (depends on 7)

The method of claim 7, wherein training the machine learning model comprises:
receiving training images;
identifying one or more objects included in the training images;
identifying one or more properties associated with each of the one or more objects in training images; and
generating a training data set including the identified objects and associated properties within the training images.

### Claim 9 (depends on 8)

The method of claim 8, wherein training the machine learning model further comprises:
receiving the training data set at the model; and
generating a property weight associated for each of the properties.

### Claim 10 (depends on 9)

The method of claim 9, wherein a training set is generated for each of a plurality of languages.

### Claim 11

A non-transitory computer-readable medium having stored thereon instructions that, when executed by one or more processors, are configurable to cause the one or more processors to:
receive an image;
identify one or more objects included in the image;
identify one or more properties associated with each of the one or more objects;
generate a matrix including data including the identified objects and associated properties; and
process the matrix at a machine learning model to determine whether the image is to be translated based on a determination that one or more objects and associated properties within the image are required to be translated.

### Claim 12 (depends on 11)

The non-transitory computer-readable medium of claim 11, wherein processing the image at the machine learning model comprises generating a property weight associated with each property.

### Claim 13 (depends on 12)

The non-transitory computer-readable medium of claim 12, wherein processing the image at the machine learning model further comprises:
determining, based on the property weights, one or more objects in the image that require translation; and
generating a list of properties associated with each of the one or more objects that require translation; and
determining, based on the property weights associated with the list of properties, whether the image is to be translated.

### Claim 14 (depends on 13)

The non-transitory computer-readable medium of claim 13, having stored thereon instructions that, when executed by one or more processors, are configurable to cause the one or more processors to train the machine learning model.

### Claim 15 (depends on 12)

The non-transitory computer-readable medium of claim 12, wherein training the machine learning model comprises:
receiving training images;
identifying one or more objects included in the training images;
identifying one or more properties associated with each of the one or more objects in training images; and
generating a training data set including the identified objects and associated properties within the training images.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein training the machine learning model comprises:
receiving the training data set at the model; and
generating a property weight associated for each of the properties.

### Claim 17

A computing device comprising:
at least one physical memory device to store image processing logic; and
one or more processors coupled with the at least one physical memory device, the one or more processors configurable to execute the image processing logic to
receive an image;
identify one or more objects included in the image;
identify one or more properties associated with each of the one or more objects;
generate a matrix including data including the identified objects and associated properties; and
process the matrix at a machine learning model to determine whether the image is to be translated based on a determination that one or more objects and associated properties within the image are required to be translated.

### Claim 18 (depends on 17)

The computing device of claim 17, wherein processing the image at the machine learning model comprises generating a property weight associated with each property.

### Claim 19 (depends on 18)

The computing device of claim 18, wherein processing the image at the machine learning model further comprises:
determining, based on the property weights, one or more objects in the image that require translation; and
generating a list of properties associated with each of the one or more objects that require translation.

### Claim 20 (depends on 19)

The computing device of claim 19, wherein an object requires translation upon a determination that an average of property weights of the object is greater than an average weight threshold value or an individual property weight is greater than an individual weight threshold value.
