patent.fyi
Applications  /  US 2026/0354949
United States Patent Application Publication

US 2026/0354949 A1

Published
Nov 12, 2026
Filed
Apr 30, 2026
Earliest date on the face
May 6, 2025
Classification
A61C 13/0004
First published here
Nov 19, 2026

Generated automatically. Not reviewed by a person. Not legal advice or an opinion.

Section 1 · machine-generated

What if?

Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.

1.1Live animated facial model

What ifWhat if the system uses a live stream of facial motion parameters to animate a face model?
Claim language“a digital video recording”

A tracking sensor outputs timestamped head pose, jaw angle and lip control values. A renderer applies the values to a face mesh built from a separate static capture and displays the dental model relative to the animated mouth.

1.2Dental casts scanned outside

What ifWhat if the system obtains the patient's tooth geometry by scanning casts on a bench?
Claim language“digital three-dimensional intraoral scan”

Elastomeric impressions form negative molds of the upper and lower dental arches, and dental stone forms positive casts. A benchtop structured-light scanner digitizes the casts, and a bite record establishes the relative positions of the arches.

1.3Continuous tracking across frames

What ifWhat if the processor keeps the video in one sequence and estimates a dental pose for every frame?
Claim language“identify and separate”

A landmark detector assigns continuous confidence weights to image measurements, and a temporal smoother estimates the dental pose across the sequence. In intervals with weak measurements, the estimate draws more heavily on nearby poses and a smooth-motion model.

1.4Calibrated dental marker tracking

What ifWhat if the system aligns the tooth model through a tracked marker fixed to an upper-arch splint?
Claim language“facial features visible in both”

A calibration scan establishes the rigid transform between the upper teeth and a marker cluster attached to the splint. The video camera estimates the cluster pose from its known point geometry, and the processor applies the calibration transform to place the upper tooth model in each usable frame.

1.5Synchronized side-by-side video

What ifWhat if the video shows the aligned dental model in a second panel next to the face footage?
Claim language“superimposed on each of the good frames”

The renderer applies the estimated dental pose and camera perspective at each retained timestamp. The encoder places the face image and the matching dental render in separate panels within the same output frame.

Section 2 · machine-generated

Variations

Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.

Other field of use

Dental overlays for animation review

A facial performance camera records an actor reading dialogue, and an intraoral scanner captures the actor's dentition. The processing software retains frames whose timestamps fall within marked dialogue takes and separates the lead-in and lead-out footage. It registers the scan to the upper teeth through their projected contours and composites the mesh over the retained footage for review of an animated character's mouth motion.

Component substitution

Contour registration with chamfer distance

A chamfer-distance optimizer replaces point-pair registration between the intraoral mesh and the facial video. For each retained frame, the software extracts tooth boundaries within the mouth region and computes a distance transform from those boundaries. The optimizer projects visible mesh contours through a calibrated camera model and adjusts the mesh pose to minimize their mean distance to the observed boundaries. A renderer overlays the mesh at the fitted pose.

Parameter or scale variation

High speed video for jaw opening

A camera records a 3-second jaw-opening sequence at 120 frames per second with a 1-millisecond exposure. The frame selector retains frames in which the tracked mouth region stays at least 40 pixels from all image edges. The software registers the intraoral mesh to the retained frames in timestamp order, using the preceding fitted pose to initialize each fit. It encodes the composite at 30 frames per second, with each retained source frame appearing once.

Combination with existing systems

Composite video with synchronized articulator

A virtual dental articulator imports separate upper and lower meshes together with their scanned bite relationship. The video pipeline registers the upper mesh to the retained facial frames and records each fitted camera pose with its frame timestamp. A jaw tracker supplies timestamped lower-jaw poses relative to the upper jaw, and the articulator interpolates those poses at the video timestamps. The renderer uses the fitted camera poses and interpolated jaw poses to display both arches in the composite video.

Automated or data-driven variant

Learned masking for natural lip occlusion

A trained segmentation model labels the lips, teeth, and mouth opening in each retained facial frame. The registration routine uses the detected tooth boundaries to fit the intraoral mesh and renders it into a separate image layer. The compositor clips the rendered layer to the mouth opening and places the original lip pixels over it so the lips cover the mesh as the mouth closes.

Joint pose fitting across video frames

A contour tracker follows visible upper tooth edges across consecutive retained frames. The software initializes the scan pose at the start of each contiguous segment and jointly refines the poses over overlapping 15-frame windows. The objective combines the distance between projected and tracked edges with a penalty on frame-to-frame changes in pose velocity. The renderer uses the estimate from the window whose center is closest to each frame to generate the overlay.

Specific version of a part

Fixed-camera facial motion capture

A tripod-mounted color camera records the full face at 60 frames per second under diffuse lighting. A fixed exposure of 1/250 second limits motion blur as the patient speaks and smiles.

Structured-light intraoral mesh

A handheld probe projects a light pattern onto the teeth and records its deformation with an offset camera. The scanner triangulates surface points, registers overlapping views and fuses them into a triangular mesh of the dental arch.

Dental overlay with lip masking

A renderer projects the aligned tooth mesh into each good frame using the calibrated camera parameters. A lip segmentation mask clips the render at the mouth opening, and the encoder places the resulting frames in timestamp order.

Section 3 · quoted from USPTO records

Other work

Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.

Document AUS 2016/0212504 A1
Published
Jul 21, 2016
Filed
Mar 28, 2016
Class
H04N 21/8545
Document BUS 2023/0218372 A1
Published
Jul 13, 2023
Filed
Mar 3, 2023
Class
A61C 7/002
Claim 1
Document A
Document B
1.1obtaining a digital video recording of a patient's face;
The interactive video includes the real-time video acquired by the video acquiring apparatus, agent identity information and agent real-time status information of each agent, and an operation menu for each agent. Therefore, using the interactive video, the user can not only learn agent information and the real-time status information of each agent, but also learn an actual situation of each agent, for example, whether an agent operator is at a workstation, and an action and a facial expression of an agent operator.
A · US 2016/0212504 A1 · ¶ 73
1.2obtaining a digital three-dimensional intraoral scan of a patient's mouth, the intraoral scan comprising a digital representation of a plurality of teeth;
In some embodiments, a method includes receiving an intraoral scan of the patient's teeth, and generating 3D dental model of the patient's teeth using the intraoral scan.
B · US 2023/0218372 A1 · abstract
1.3processing the digital video recording to identify and separate a plurality of good frames and one or more bad frames based on predetermined criteria;
1.4aligning the digital three-dimensional intraoral scan to the plurality of good frames based on facial features visible in both the intraoral scan and the good frames; and
1.5generating a composite video using at least the plurality of good frames, wherein the aligned digital three-dimensional intraoral scan is superimposed on each of the good frames to provide a dynamic visual representation of the patient's teeth in relation to their facial appearance.
Using the video surveillance method, and the surveillance server and system provided in embodiments of the present disclosure, real-time presentation information and an acquired real-time video are superimposed according to a superimposition position and agent management information in a form of an operation menu and the acquired real-time video are superimposed according to the superimposition position in order to generate an interactive video.
A · US 2016/0212504 A1 · ¶ 38

An empty cell means only that no passage was reproduced there.

Statements in these documents
Using the foregoing surveillance system, the surveillance server can only acquire and send the agent identity information and the agent status information to the client, and the user can only learn the agent identity information and the agent status information according to an image displayed by the client and cannot learn an actual image of the agent in real time, which may cause an inauthentic surveillance result.
A · US 2016/0212504 A1 · ¶ 4
In view of this, the present disclosure provides a video surveillance method, a surveillance server, and a surveillance system that enable a user to learn an actual image of an agent, thereby ensuring authenticity of a surveillance result.
A · US 2016/0212504 A1 · ¶ 5
Cite aspatent.fyi, US 2026/0354949 A1, claim 1, first published 2026-11-19T06:07:06-05:00, https://patent.fyi/US20260354949/, SHA-256 49c414ad768e922adddd86e4a00d1a561722af3a86dd1daf9dd6d09436ccfce9
First published
Permanent addresshttps://patent.fyi/US20260354949/
Content fingerprint (SHA-256)49c414ad768e922adddd86e4a00d1a561722af3a86dd1daf9dd6d09436ccfce9
Methodv2026-11-18.2 · automated, not reviewed by a person
Correctionscorrections@patent.fyi