IP Library Granted Patent US 12,705,775
Granted Patent B2
US 12,705,775 · App. 17/902,561 · Granted Aug 11, 2026

Method and apparatus for obtaining 3D information of vehicle

Inventors: Yiqiang Chen (Shenzhen, CN); Jie Zhu (Shenzhen, CN); Lijun Xue (Shenzhen, CN)
Assignee: Yinwang Intelligent Technologies Co., Ltd.
G06T7/70G06T17/00G06V10/24G06T2207/30252G06T2210/12H04W4/40
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,705,775
App. No.
17/902,561
Granted
Aug 11, 2026
Kind
B2
Abstract

A method and an apparatus for obtaining 3D information of a vehicle are provided. The method includes: first determining a body boundary line of a first vehicle, and then determining an observation angle and/or an orientation angle of the first vehicle based on the body boundary line.

Claims (236)

1 . A method for obtaining three dimensional (3D) information of a vehicle, comprising:

obtaining a body boundary line of a first vehicle, wherein the body boundary line comprises a boundary line at a boundary between a body and a head of the first vehicle, or a boundary line at a boundary between the body and a tail of the first vehicle; and

determining at least one of an observation angle of the first vehicle or an orientation angle of the first vehicle based on the body boundary line, wherein the determining the observation angle of the first vehicle based on the body boundary line comprises:

determining, based on the body boundary line and a width of a preset two dimensional (2D) bounding frame, a projection of a body length of the first vehicle or a projection of a body width of the first vehicle on a plane perpendicular to an optical center direction; and

determining the observation angle of the first vehicle based on the projection of the body length of the first vehicle or the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and a known value representing a preset actual length-to-width ratio of the first vehicle, wherein the observation angle of the first vehicle, the projection of the body length of the first vehicle or the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the known value representing the preset actual length-to-width ratio of the first vehicle meet a preset condition; and wherein

the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the preset actual length-to-width ratio of the first vehicle meet

θ

=

arc

tan

(

ω

-

a

μ

a

)

 with the observation angle of the first vehicle, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the preset actual length-to-width ratio of the first vehicle meet

θ

=

arc

tan

(

b

μ

(

ω

-

b

)

)

 with the observation angle of the first vehicle, wherein ω is the width of the preset 2D bounding box, μ is the preset actual length-to-width ratio of the first vehicle, α′ is the projection of the body width of the first vehicle in a direction parallel to an image plane, b is the projection of the body length of the first vehicle in the direction parallel to the image plane, and θ is the observation angle.

2 . The method according to claim 1 , wherein the determining the orientation angle of the first vehicle based on the body boundary line comprises:

determining, based on the body boundary line and the width of the preset 2D bounding box, the projection of the body length of the first vehicle in the direction parallel to the image plane;

determining, based on the body boundary line and the width of the preset 2D bounding box, the projection of the body width of the first vehicle in the direction parallel to the image plane; and

determining the orientation angle of the first vehicle based on the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, respective angles between two edges of the preset 2D bounding box and an optical center, and the preset actual length-to-width ratio of the first vehicle.

3 . The method according to claim 2 , wherein the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, the respective angles between the two edges of the preset 2D bounding box and the optical center, and the preset actual length-to-width ratio of the first vehicle meet

φ

=

arc

tan

(

μ

b

1

-

a

tan

β

a

+

μ

b

1

tan

α

)

with the orientation angle of the first vehicle, wherein

a is the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, b1 is the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, α is an angle between a left edge of the 2D bounding box and the optical center, β is an angle between a right edge of the 2D bounding box and the optical center, and φ is the orientation angle of the first vehicle, and μ is the preset actual length-to-width ratio of the first vehicle.

4 . The method according to claim 1 , wherein the method further comprises:

determining, based on the orientation angle of the first vehicle and the body boundary line, 3D bounding box information corresponding to the first vehicle.

5 . An apparatus, comprising:

at least one processor; and

one or more memories coupled to the at least one processor and storing programming instructions for execution by the at least one processor to cause the apparatus to:

obtain a body boundary line of a first vehicle, wherein the body boundary line comprises a boundary line at a boundary between a body and a head of the first vehicle, or a boundary line at a boundary between the body and a tail of the first vehicle; and

determine at least one of an observation angle of the first vehicle or an orientation angle of the first vehicle based on the body boundary line, wherein the determine the observation angle of the first vehicle based on the body boundary line comprises:

determine, based on the body boundary line and a width of a preset two dimensional (2D) bounding frame, a projection of a body length of the first vehicle or a projection of a body width of the first vehicle on a plane perpendicular to an optical center direction; and

determine the observation angle of the first vehicle based on the projection of the body length of the first vehicle or the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and a known value representing a preset actual length-to-width ratio of the first vehicle, wherein the observation angle of the first vehicle, the projection of the body length of the first vehicle or the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the known value representing the preset actual length-to-width ratio of the first vehicle meet a preset condition; and wherein

the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the preset actual length-to-width ratio of the first vehicle meet

θ

=

arc

tan

(

ω

-

a

μ

a

)

 with the observation angle of the first vehicle, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the preset actual length-to-width ratio of the first vehicle meet

θ

=

arc

tan

(

b

μ

(

ω

-

b

)

)

 with the observation angle of the first vehicle, wherein ω is the width of the preset 2D bounding box, μ is the preset actual length-to-width ratio of the first vehicle, a′ is the projection of the body width of the first vehicle in a direction parallel to an image plane, b is the projection of the body length of the first vehicle in the direction parallel to the image plane, and θ is the observation angle.

6 . The apparatus according to claim 5 , wherein the determine the orientation angle of the first vehicle based on the body boundary line comprises:

determine, based on the body boundary line and the width of the preset 2D bounding box, the projection of the body length of the first vehicle in the direction parallel to the image plane;

determine, based on the body boundary line and the width of the preset 2D bounding box, the projection of the body width of the first vehicle in the direction parallel to the image plane; and

determine the orientation angle of the first vehicle based on the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, respective angles between two edges of the preset 2D bounding box and an optical center, and the preset actual length-to-width ratio of the first vehicle.

7 . The apparatus according to claim 6 , wherein the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, the respective angles between the two edges of the preset 2D bounding box and the optical center, and the preset actual length-to-width ratio of the first vehicle meet

φ

=

arc

tan

(

μ

b

1

-

a

tan

β

a

+

μ

b

1

tan

α

)

with the orientation angle of the first vehicle, wherein

a is the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, b1 is the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, α is an angle between a left edge of the 2D bounding box and the optical center, β is an angle between a right edge of the 2D bounding box and the optical center, and φ is the orientation angle of the first vehicle, and μ is the preset actual length-to-width ratio of the first vehicle.

8 . The apparatus according to claim 5 , wherein the programming instructions, when executed by the at least one processor, cause the apparatus to:

determine, based on the orientation angle of the first vehicle and the body boundary line, 3D bounding box information corresponding to the first vehicle.

9 . One or more non-transitory computer-readable media storing computer instructions, that when executed by one or more processors, cause a computing device to perform operations comprising:

obtaining a body boundary line of a first vehicle, wherein the body boundary line comprises a boundary line at a boundary between a body and a head of the first vehicle, or a boundary line at a boundary between the body and a tail of the first vehicle; and

determining at least one of an observation angle of the first vehicle or an orientation angle of the first vehicle based on the body boundary line, wherein the determining the observation angle of the first vehicle based on the body boundary line comprises:

determining, based on the body boundary line and a width of a preset two dimensional (2D) bounding frame, a projection of a body length of the first vehicle or a projection of a body width of the first vehicle on a plane perpendicular to an optical center direction; and

determining the observation angle of the first vehicle based on the projection of the body length of the first vehicle or the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and a known value representing a preset actual length-to-width ratio of the first vehicle, wherein the observation angle of the first vehicle, the projection of the body length of the first vehicle or the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the known value representing the preset actual length-to-width ratio of the first vehicle meet a preset condition; and wherein

the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the preset actual length-to-width ratio of the first vehicle meet

φ

=

arc

tan

(

ω

-

a

μ

a

)

 with the observation angle of the first vehicle, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, the width of the preset 2D bounding box, and the preset actual length-to-width ratio of the first vehicle meet

θ

=

arc

tan

(

b

μ

(

ω

-

b

)

)

 with the observation angle of the first vehicle, wherein ω is the width of the preset 2D bounding box, μ is the preset actual length-to-width ratio of the first vehicle, a′ is the projection of the body width of the first vehicle in a direction parallel to an image plane, b is the projection of the body length of the first vehicle in the direction parallel to the image plane, and θ is the observation angle.

10 . The one or more non-transitory computer-readable media according to claim 9 , wherein the determining the orientation angle of the first vehicle based on the body boundary line comprises:

determining, based on the body boundary line and the width of the preset 2D bounding box, the projection of the body length of the first vehicle in the direction parallel to the image plane;

determining, based on the body boundary line and the width of the preset 2D bounding box, the projection of the body width of the first vehicle in the direction parallel to the image plane; and

determining the orientation angle of the first vehicle based on the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, respective angles between two edges of the preset 2D bounding box and an optical center, and the preset actual length-to-width ratio of the first vehicle.

11 . The one or more non-transitory computer-readable media according to claim 10 , wherein the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, the respective angles between the two edges of the preset 2D bounding box and the optical center, and the preset actual length-to-width ratio of the first vehicle meet

φ

=

arctan

(

μ

b

1

-

a

tan

β

a

+

μ

b

1

tan

α

)

with the orientation angle of the first vehicle, wherein

a is the projection of the body length of the first vehicle on the plane perpendicular to the optical center direction, b1 is the projection of the body width of the first vehicle on the plane perpendicular to the optical center direction, α is an angle between a left edge of the 2D bounding box and the optical center, β is an angle between a right edge of the 2D bounding box and the optical center, and φ is the orientation angle of the first vehicle, and μ is the preset actual length-to-width ratio of the first vehicle.

12 . The one or more non-transitory computer-readable media according to claim 9 , wherein the operations comprise: determining, based on the orientation angle of the first vehicle and the body boundary line, 3D bounding box information corresponding to the first vehicle.

Assignments (3)
CHANGE OF NAME Recorded Apr 28, 2026
From: SHENZHEN YINWANG INTELLIGENT TECHNOLOGIES CO., LTD.
To: YINWANG INTELLIGENT TECHNOLOGIES CO., LTD.
Reel/Frame 075492/0796 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 12, 2024
From: HUAWEI TECHNOLOGIES CO., LTD.
To: SHENZHEN YINWANG INTELLIGENT TECHNOLOGIES CO., LTD.
Reel/Frame 069335/0897 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 5, 2024
From: CHEN, YIQIANG; ZHU, JIE; XUE, LIJUN
To: HUAWEI TECHNOLOGIES CO., LTD.
Reel/Frame 066036/0608 →
Priority Claims (1)
CN 202010152507.4 · Mar 6, 2020 · national
Continuity (2)
Continuation PCTCN2021076863 · Feb 19, 2021
Related Publication 20220414917A1 · Dec 29, 2022
References Cited (21)
US 9533575B2 · Tran et al. · 2017 [cited by applicant]
US 11024045B2 · Wang · 2021 [cited by examiner]
CN 102930242A · 2013 [cited by applicant]
CN 107093210A · 2017 [cited by applicant]
CN 109242903A · 2019 [cited by applicant]
CN 109631896A · 2019 [cited by applicant]
CN 110414355A · 2019 [cited by applicant]
CN 110427797A · 2019 [cited by applicant]
CN 110517349A · 2019 [cited by applicant]
CN 111081033A · 2020 [cited by applicant]
DE 102019002269A1 · 2020 [cited by applicant]
EP 3621036A1 · 2020 [cited by applicant]
JP 2001128158A · 2001 [cited by applicant]
Bazan et al., “A new method to estimate dimensions of vehicle using a single camera,” Archives of Transport System Telematics, vol. 10, Issue 3, Sep. 2017, available at: https://bibliotekanauki.pl/articles/393575.pdf (l… [cited by examiner]
Fang et al., “3D Bounding Box Estimation for Autonomous Vehicles by Cascaded Geometric Constraints and Depurated 2D Detections Using 3D Results,” arXiv: 1909.01867v1 [cs.CV], Sep. 1, 2019, https://doi.org/10.48550/arXiv… [cited by examiner]
Schoepflin et al., “Dynamic camera calibration of roadside traffic management cameras for vehicle speed estimation,” in IEEE Transactions on Intelligent Transportation Systems, vol. 4, No. 2, pp. 90-98, Jun. 2003, doi: … [cited by examiner]
Extended European Search Report in European Appln No. 21765079.5, dated Jun. 28, 2023, 9 pages. [cited by applicant]
Mousavian et al., “3D Bounding Box Estimation Using Deep Learning and Geometry,” Submitted on Apr. 10, 2017, arXiv:1612.00496v2 [cs.CV] , 10 pages. [cited by applicant]
Qi et al., “PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation,” Submitted on Apr. 10, 2017, arXiv:1612.00593v2 [cs.CV], 19 pages. [cited by applicant]
Zhou et al., “VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection,” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun. 18-23, 2018, 10 pages. [cited by applicant]
International Search Report and Written Opinion in International Appln. No. PCT/CN2021/076863, mailed on May 17, 2021, 18 pages (with English translation). [cited by applicant]