# US 2024/0203120 A1 · Generating mapping information based on image locations

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

- Number: US 2024/0203120 A1
- Document ID: US20240203120A1
- Status: Published application
- Kind: A1
- Application number: 18/585,663
- Filed: 2024-02-23
- Published: 2024-06-20
- Earliest claimed: 2020-10-20
- Assignees: DoorDash, Inc.
- Applicants: Doordash, Inc.
- Inventors: Sushil VELLANKI; Kuleen NIMKAR; Li Xiang TIAN
- Web page: https://patentgrep.com/patent/US20240203120A1

## CPC

- G06V 20/20: in augmented reality scenes
- G06V 20/70: Labelling scene content, e.g. deriving syntactic or semantic representations
- G01C 21/3667: {Display of a road map (G01C21/3614 takes precedence; guidance using 3D or perspective road maps G01C21/3635)}
- G06N 20/00: Machine learning
- G06T 7/74: {involving reference images or patches}
- G06V 20/176: {Urban or other man-made structures}
- G06V 20/38: {Outdoor scenes}
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30184: Infrastructure
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

In some examples, a system may receive over time, from one or more agent devices, and in association with a delivery location, a plurality of images and associated respective location data. Further, the respective location data associated with at least one of the images can differ from the respective location data associated with at least one other one of the images. The plurality of images are input to a machine-learning model that is trained to determine whether individual images include a threshold amount of information. Based at least on the machine-learning model indicating that the individual images satisfy the threshold amount of information, the system determines, based on at least one of averaging or clustering of the respective location information associated with the plurality of images, a consensus location for the delivery location. The system stores the consensus location information as mapping information associated with the delivery location.

## Claims

### Claim 1

A system comprising:
one or more processors; and
one or more computer-readable media storing instructions executable to configure the one or more processors to perform operations including:
receiving, by the one or more processors, over time, from one or more agent devices and in association with a delivery location, a plurality of images and associated respective location data, the respective location data associated with at least one of the images of the plurality of images differing from the respective location data associated with at least one other one of the images of the plurality of images;
providing the plurality of images as inputs to a machine-learning model that is trained to determine whether individual images of the plurality of images include a threshold amount of information;
based at least in part on the machine-learning model indicating that the individual images of the plurality of received images satisfy the threshold amount of information, determining, based on at least one of averaging or clustering of the respective location information associated with the plurality of images, a consensus location for the delivery location; and
storing the consensus location information as mapping information associated with the delivery location.

### Claim 2 (depends on 1)

The system as recited in claim 1, the operations further comprising:
receiving a request for an item for delivery to the delivery location; and
retrieving the consensus location as the mapping information associated with the delivery location for use in generating a map to present on an agent device for delivery of the item to the delivery location.

### Claim 3 (depends on 1)

The system as recited in claim 1, the operations further comprising:
receiving feedback indicating that an item associated with one of the received images of the plurality of images was not received or was in a wrong location; and
removing the associated respective location data from being associated with the delivery location when determining the consensus location.

### Claim 4 (depends on 1)

The system as recited in claim 1, the operations further comprising:
receiving another image and associated respective location data for a different delivery location;
providing the other image as input to the machine-learning model; and
based at least in part on an output of the machine-learning model indicating that the other image fails to satisfy the threshold amount of information, sending, by the one or more processors, to an agent device that sent the other image, an instruction to capture an additional image corresponding to the different delivery location.

### Claim 5 (depends on 1)

The system as recited in claim 1, the operations further comprising:
receiving another image and associated respective location data for a different delivery location;
providing the other image as input to the machine-learning model; and
based at least in part on an output of the machine-learning model indicating that the other image fails to satisfy the threshold amount of information, excluding the respective location data associated with the other received image from being associated with mapping information for the different delivery location.

### Claim 6 (depends on 1)

The system as recited in claim 1, the operations further comprising training the machine learning model using a plurality of images of past delivery locations for a plurality of past deliveries to densely populated structures.

### Claim 7 (depends on 1)

The system as recited in claim 1, wherein the threshold amount of information includes a delivered item and at least one of an entrance portion, a door portion, or a unit number.

### Claim 8

A method comprising:
receiving, by one or more processors, over time, from one or more agent devices and in association with a delivery location, a plurality of images and associated respective location data, the respective location data associated with at least one of the images of the plurality of images differing from the respective location data associated with at least one other one of the images of the plurality of images;
providing, by the one or more processors, the plurality of images as inputs to a machine-learning model that is trained to determine whether individual images of the plurality of images include a threshold amount of information;
based at least in part on the machine-learning model indicating that the individual images of the plurality of received images satisfy the threshold amount of information, determining, by the one or more processors, based on at least one of averaging or clustering of the respective location information associated with the plurality of images, a consensus location for the delivery location; and
storing, by the one or more processors, the consensus location information as mapping information associated with the delivery location.

### Claim 9 (depends on 8)

The method as recited in claim 8, further comprising:
receiving a request for an item for delivery to the delivery location; and
retrieving the consensus location as the mapping information associated with the delivery location for use in generating a map to present on an agent device for delivery of the item to the delivery location.

### Claim 10 (depends on 8)

The method as recited in claim 8, further comprising:
receiving feedback indicating that an item associated with one of the received images of the plurality of images was not received or was in a wrong location; and
removing the associated respective location data from being associated with the delivery location when determining the consensus location.

### Claim 11 (depends on 8)

The method as recited in claim 8, further comprising:
receiving another image and associated respective location data for a different delivery location;
providing the other image as input to the machine-learning model; and
based at least in part on an output of the machine-learning model indicating that the other image fails to satisfy the threshold amount of information, sending, by the one or more processors, to an agent device that sent the other image, an instruction to capture an additional image corresponding to the different delivery location.

### Claim 12 (depends on 8)

The method as recited in claim 8, further comprising:
receiving another image and associated respective location data for a different delivery location;
providing the other image as input to the machine-learning model; and
based at least in part on an output of the machine-learning model indicating that the other image fails to satisfy the threshold amount of information, excluding the respective location data associated with the other received image from being associated with mapping information for the different delivery location.

### Claim 13 (depends on 8)

The method as recited in claim 8, further comprising training the machine learning model using a plurality of images of past delivery locations for a plurality of past deliveries to densely populated structures.

### Claim 14 (depends on 8)

The method as recited in claim 8, wherein the threshold amount of information includes a delivered item and at least one of an entrance portion, a door portion, or a unit number.

### Claim 15

A non-transitory computer-readable medium maintaining instructions executable to configure one or more processors to perform operations comprising:
receiving, by the one or more processors, over time, from one or more agent devices and in association with a delivery location, a plurality of images and associated respective location data, the respective location data associated with at least one of the images of the plurality of images differing from the respective location data associated with at least one other one of the images of the plurality of images;
providing the plurality of images as inputs to a machine-learning model that is trained to determine whether individual images of the plurality of images include a threshold amount of information;
based at least in part on the machine-learning model indicating that the individual images of the plurality of received images satisfy the threshold amount of information, determining, based on at least one of averaging or clustering of the respective location information associated with the plurality of images, a consensus location for the delivery location; and
storing the consensus location information as mapping information associated with the delivery location.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium as recited in claim 15, the operations further comprising:
receiving a request for an item for delivery to the delivery location; and
retrieving the consensus location as the mapping information associated with the delivery location for use in generating a map to present on an agent device for delivery of the item to the delivery location.

### Claim 17 (depends on 15)

The non-transitory computer-readable medium as recited in claim 15, the operations further comprising:
receiving feedback indicating that an item associated with one of the received images of the plurality of images was not received or was in a wrong location; and
removing the associated respective location data from being associated with the delivery location when determining the consensus location.

### Claim 18 (depends on 15)

The non-transitory computer-readable medium as recited in claim 15, the operations further comprising:
receiving another image and associated respective location data for a different delivery location;
providing the other image as input to the machine-learning model; and
based at least in part on an output of the machine-learning model indicating that the other image fails to satisfy the threshold amount of information, sending, by the one or more processors, to an agent device that sent the other image, an instruction to capture an additional image corresponding to the different delivery location.

### Claim 19 (depends on 15)

The non-transitory computer-readable medium as recited in claim 15, the operations further comprising:
receiving another image and associated respective location data for a different delivery location;
providing the other image as input to the machine-learning model; and
based at least in part on an output of the machine-learning model indicating that the other image fails to satisfy the threshold amount of information, excluding the respective location data associated with the other received image from being associated with mapping information for the different delivery location.

### Claim 20 (depends on 15)

The non-transitory computer-readable medium as recited in claim 15, the operations further comprising training the machine learning model using a plurality of images of past delivery locations for a plurality of past deliveries to densely populated structures.
