# US 2025/0220128 A1 · Panorama Generation with Mobile Camera

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

- Number: US 2025/0220128 A1
- Document ID: US20250220128A1
- Status: Published application
- Kind: A1
- Application number: 19/084,090
- Filed: 2025-03-19
- Published: 2025-07-03
- Earliest claimed: 2022-09-19
- Assignees: Google Inc.
- Applicants: Google LLC
- Inventors: Lin Chen; Wei (Alex) Hong
- Web page: https://patentgrep.com/patent/US20250220128A1

## CPC

- H04N 5/265: Mixing
- G06T 7/0002: {Inspection of images, e.g. flaw detection}
- G06V 10/16: {using multiple overlapping images; Image stitching}
- G06V 10/25: Determination of region of interest [ROI] or a volume of interest [VOI]
- G06V 40/161: {Detection; Localisation; Normalisation}
- H04N 23/698: for achieving an enlarged field of view, e.g. panoramic image capture
- G06T 2207/20212: Image combination
- G06T 2207/30168: Image quality inspection
- G06T 2207/30201: Face
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Devices and methods for selecting and stitching image frames are provided. A method includes obtaining a plurality of image frames. The method also includes identifying one or more regions of interest within one or more image frames in the plurality of image frames. The method further includes selecting, based on a respective quality measure associated with each image frame of the plurality of image frames, a set of base frames, where each identified region of interest of the one or more identified regions of interest is fully contained within at least one base frame in the selected set of base frames. The method additionally includes stitching together the selected set of base frames to create a composite image.

## Claims

### Claim 1

A computer-implemented method comprising:
obtaining, by a computing device, a plurality of image frames;
identifying, by the computing device, one or more regions of interest (ROI) within one or more image frames in the plurality of image frames;
associating each identified ROI within the one or more image frames with a respective identifier, wherein two different ROIs are associated with two different identifiers;
selecting, by the computing device, a set of base frames, wherein the selecting of the set of base frames comprises:
comparing one or more current identifiers in a current frame with previous identifiers in previous frames to determine that a current ROI in the current frame does not appear in the previous frames, wherein the previous frames are subsequent to a previously selected base frame, and
upon a determination that the current ROI in the current frame does not appear in the previous frames, adding the current frame to the set of base frames; and
stitching together, by the computing device, the selected set of base frames to create a composite image.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein each of the one or more regions of interest corresponds to a region containing a face.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the selecting of the set of base frames is based on a respective quality measure associated with each image frame of the plurality of image frames.

### Claim 4 (depends on 3)

The computer-implemented method of claim 3, wherein the respective quality measure associated with each image frame of the plurality of image frames comprises a metric based on at least one of: motion blur of the image frame, focus of the image frame, or exposure of the image frame.

### Claim 5 (depends on 3)

The computer-implemented method of claim 3, wherein the respective quality measure is computationally biased to give greater weight to pixels that are located within the one or more identified regions of interest.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein selecting the set of base frames comprises:
determining, from the plurality of image frames, a plurality of subsets of image frames, wherein each of the subsets comprises image frames that contain the same one or more regions of interest; and
selecting, from each of the subsets and based on the respective quality measure associated with each image frame in the subset, a base frame.

### Claim 7 (depends on 6)

The computer-implemented method of claim 6, wherein the selecting of the base frame from each of the subsets comprises selecting an image frame from amongst the image frames of the subset, wherein a quality measure associated with the selected base frame exceeds a quality threshold.

### Claim 8 (depends on 6)

The computer-implemented method of claim 6, wherein the identifying of the one or more regions of interest comprises determining unique identifiers for each of the one or more regions of interest, and wherein each of the subsets comprises one or more image frames that contain the same unique identifiers.

### Claim 9 (depends on 8)

The computer-implemented method of claim 8, wherein each image frame of the plurality of image frames was captured by a camera device at a respective rotation angle, and wherein determining a subset of image frames comprises:
ordering the plurality of image frames based on the respective rotation angle;
designating a starting image frame from the plurality of image frames; and
beginning from the starting image frame, iterating through the plurality of image frames according to the ordering until the next image frame to be iterated through is an image frame with at least one different unique identifier than the starting image frame, wherein the subset of image frames comprises the image frames that were iterated through.

### Claim 10 (depends on 9)

The computer-implemented method of claim 9, wherein the rotation angle comprises a measure based on a horizontal angle of the camera device.

### Claim 11 (depends on 9)

The computer-implemented method of claim 9, wherein designating the starting image frame comprises designating a first image frame from the ordering.

### Claim 12 (depends on 9)

The computer-implemented method of claim 9, wherein the subset of image frames is a first subset of image frames, and wherein determining a second subset of image frames comprises:
designating a second starting image frame from the plurality of image frames; and
beginning from the second starting image frame, iterating through the plurality of image frames according to the ordering until the next image frame to be iterated through is an image frame with at least one different unique identifier than the second starting image frame, and
wherein the second subset of image frames comprises the image frames that were iterated through beginning from the second starting image frame.

### Claim 13 (depends on 12)

The computer-implemented method of claim 12, wherein the designating of the second starting image frame comprises designating the image frame with at least one different unique facial identifier than the starting image frame.

### Claim 14 (depends on 9)

The computer-implemented method of claim 9, wherein the stitching occurs after each image frame from the plurality of image frames has been iterated through at least once.

### Claim 15 (depends on 1)

The computer-implemented method of claim 1, wherein the selecting of the set of base frames is performed so that each identified region of interest of the one or more identified regions of interest is fully contained within at least one base frame in the selected set of base frames.

### Claim 16 (depends on 1)

The computer-implemented method of claim 1, wherein the stitching comprises a seam finding process that involves adding a computational bias to seams that contain pixels from the one or more identified regions of interest.

### Claim 17 (depends on 16)

The computer-implemented method of claim 16, wherein the computational bias involves adding a penalty term to any seam that contains pixels from the one or more identified regions of interest.

### Claim 18 (depends on 1)

The computer-implemented method of claim 1, further comprising an optical flow seam repair step that involves:
determining overlapping regions for the selected of set of base frames;
computing respective optical flow fields for each of the overlapping regions; and
applying the computed optical flow fields to align the overlapping regions.

### Claim 19 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of image frames are captured by a camera device in one continuous stream.

### Claim 20 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of image frames are captured using a front facing camera of a camera device.

### Claim 21 (depends on 1)

The computer-implemented method of claim 1, further comprising:
comparing one or more second current identifiers in a second current frame with second previous identifiers associated with second previous frames to determine that ROIs in the second current frame appear in the second previous frames, wherein the second previous frames are subsequent to a previously selected second base frame; and
upon a determination that the ROIs in the second current frame appear in the second previous frames, not adding the second current frame to the set of base frames.

### Claim 22

A computing device, comprising:
one or more processors; and
non-transitory data storage storing at least computer-readable instructions that, when executed by the one or more processors, cause the computing device to perform operations comprising:
obtaining a plurality of image frames;
identifying one or more regions of interest (ROI) within one or more image frames in the plurality of image frames;
associating each identified ROI within the one or more image frames with a respective identifier, wherein two different ROIs are associated with two different identifiers;
selecting, by the computing device, a set of base frames, wherein the selecting of the set of base frames further comprising:
comparing one or more current identifiers in a current frame with previous identifiers in previous frames to determine that a current ROI in the current frame does not appear in the previous frames, wherein the previous frames are subsequent to a previously selected base frame, and
upon a determination that the current ROI in the current frame does not appear in the previous frames, adding the current frame to the set of base frames; and
stitching together the selected set of base frames to create a composite image.

### Claim 23

An article of manufacture comprising non-transitory data storage storing at least computer-readable instructions that, when executed by one or more processors of a computing device, cause the computing device to perform operations comprising:
obtaining a plurality of image frames;
identifying one or more regions of interest (ROI) within one or more image frames in the plurality of image frames;
associating each identified ROI within the one or more image frames with a respective identifier, wherein two different ROIs are associated with two different identifiers;
selecting, by the computing device, a set of base frames, wherein the selecting of the set of base frames further comprising:
comparing one or more current identifiers in a current frame with previous identifiers in previous frames to determine that a current ROI in the current frame does not appear in the previous frames, wherein the previous frames are subsequent to a previously selected base frame, and
upon a determination that the current ROI in the current frame does not appear in the previous frames, adding the current frame to the set of base frames; and
stitching together the selected set of base frames to create a composite image.
