IP Library Granted Patent US 12,380,693
Granted Patent B2
US 12,380,693 · App. 17/635,385 · Granted Aug 5, 2025

Object detection circuitry and object detection method using combined probability data

Inventor: Florian Becker (Stuttgart, DE)
Assignee: SONY GROUP CORPORATION
G06V20/00G06T7/246G06T7/74G06V10/7715G06V10/776G06V10/806G06T2207/10028G06T2207/10044G06T2207/20081G06V2201/07
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 12,380,693
App. No.
17/635,385
Granted
Aug 5, 2025
Kind
B2
Abstract

The present disclosure generally pertains to an object detection circuitry configured to: obtain first feature data which are based on first sensing data of a first sensor; compare the first feature data to a first predetermined feature model being representative of a predefined object, wherein the first predetermined feature model is specific for the first sensor, thereby generating first object probability data; obtain second feature data which are based on second sensing data of a second sensor; compare the second feature data to a second predetermined feature model being representative of the predefined object, wherein the second predetermined feature model is specific for the second sensor, thereby generating second object probability data; and combine the first and the second object probability data, thereby generating combined probability data for detecting the predefined object.

Claims (52)

1. An object detection circuitry configured to:

obtain first feature data which are based on first sensing data of a first sensor;

compare the first feature data to a first predetermined feature model being representative of a predefined object O thereby generating first object probability data representing a first probability of an existence of the predefined object O in a first object parameter space, wherein the first predetermined feature model represents a plurality of different postures of the predefined object O and is specific for the first sensor;

obtain second feature data which are based on second sensing data of a second sensor;

compare the second feature data to a second predetermined feature model being representative of the predefined object O thereby generating second object probability data representing a second probability of an existence of the predefined object O in a second object parameter space, wherein the second predetermined feature model represents a plurality of different postures of the predefined object O and is specific for the second sensor;

combine the first and the second object probability data, thereby generating combined probability data for detecting the predefined object O; and

detect the predefined object O using the combined probability data.

2. The object detection circuitry of claim 1 , further configured to:

detect a plurality of maxima of the combined probability data being indicative for at least one position parameter of the predefined object O; and

determine the at least one position parameter of the predefined object O.

3. The object detection circuitry of claim 2 , wherein the at least one position parameter includes at least one of a position, a distance, an angle, or a posture.

4. The object detection circuitry of claim 2 , further configured to:

determine at least one of a correctness or a precision of the detection of the predefined object O.

5. The object detection circuitry of claim 2 , further configured to:

track the detected predefined object O.

6. The object detection circuitry of claim 1 , further configured to:

generate a first feature map based on the first sensing data;

generate a second feature map based on the second sensing data; and

transfer the first and the second feature map into a predefined coordinate system.

7. The object detection circuitry of claim 1 , wherein the first and the second sensor include at least one of a radar sensor, a lidar sensor, a camera, or a time-of-flight sensor.

8. The object detection circuitry of claim 1 , wherein at least one of the first predetermined feature model or the second predetermined feature model is based on a supervised training of an artificial intelligence.

9. The object detection circuitry of claim 1 , wherein the predefined object O is based on a class of predefined objects.

10. The object detection circuitry of claim 1 , further configured to:

iteratively convolve the first feature data with the first predetermined feature model, thereby generating the first object probability data;

iteratively convolve the second feature data with the second predetermined feature model, thereby generating the second object probability data; and

iteratively convolve the first and the second object probability data, thereby generating the combined object probability data.

11. An object detection method comprising:

obtaining first feature data which are based on first sensing data of a first sensor;

comparing the first feature data to a first predetermined feature model being representative of a predefined object O thereby generating first object probability data representing a first probability of an existence of the predefined object O in a first object parameter space, wherein the first predetermined feature model represents a plurality of different postures of the predefined object O and is specific for the first sensor;

obtaining second feature data which are based on second sensing data of a second sensor;

comparing the second feature data to a second predetermined feature model being representative of the predefined object O thereby generating second object probability data representing a second probability of an existence of the predefined object O in a second object parameter space, wherein the second predetermined feature model represents a plurality of different postures of the predefined object O and is specific for the second sensor;

combining the first and the second object probability data, thereby generating combined probability data for detecting the predefined object O; and

detecting the predefined object O using the combined probability data.

12. The object detection method of claim 11 , further comprising:

detecting a plurality of maxima of the combined probability data being indicative for at least one position parameter of the predefined object O; and

determining the at least one position parameter of the predefined object O.

13. The object detection method of claim 12 , wherein the at least one position parameter includes at least one of a position, a distance, an angle, or a posture.

14. The object detection method of claim 12 , further comprising:

determining at least one of a correctness or a precision of the detection of the predefined object O.

15. The object detection method of claim 12 , further comprising:

tracking the detected predefined object O.

16. The object detection method of claim 11 , further comprising:

generating a first feature map based on the first sensing data;

generating a second feature map based on the second sensing data; and

transferring the first and the second feature map into a predefined coordinate system.

17. The object detection method of claim 11 , wherein the first and the second sensor include at least one of a radar sensor, a lidar sensor, a camera, or a time-of-flight sensor.

18. The object detection method of claim 11 , wherein at least one of the first predetermined feature model or the second predetermined feature model is based on a supervised training of an artificial intelligence.

19. The object detection method of claim 11 , wherein the predefined object O is based on a class of predefined objects.

20. The object detection method of claim 11 , further comprising:

iteratively convolving the first feature data with the first predetermined feature model, thereby generating the first object probability data;

iteratively convolving the second feature data with the second predetermined feature model, thereby generating the second object probability data; and

iteratively convolving the first and the second object probability data, thereby generating the combined probability data.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 15, 2022
From: BECKER, FLORIAN
To: SONY GROUP CORPORATION
Reel/Frame 059013/0213 →
Priority Claims (1)
EP 19199664 · Sep 25, 2019 · regional
Continuity (1)
Related Publication 20220406044A1 · Dec 22, 2022
References Cited (19)
US 9684836B1 · Ferguson et al. · 2017 [cited by applicant]
US 11062454B1 · Cohen · 2021 [cited by examiner]
US 20160299229A1 · Heinonen · 2016 [cited by examiner]
US 20180068206A1 · Pollach et al. · 2018 [cited by applicant]
US 20180365888A1 · Satzoda · 2018 [cited by examiner]
US 20190026917A1 · Liao · 2019 [cited by examiner]
US 20190279000A1 · Pfeifle · 2019 [cited by examiner]
US 20200280429A1 · Su · 2020 [cited by examiner]
JP 2017146957A · 2017 [cited by applicant]
WO WO2017057056A1 · 2017 [cited by applicant]
Shi, J., Wang, W., Wang, X., Sun, H., Lan, X., Xin, J., & Zheng, N. (Jun. 2018). Leveraging spatio-temporal evidence and independent vision channel to improve multi-sensor fusion for vehicle environmental perception. In… [cited by examiner]
International Search Report and Written Opinion mailed on Dec. 7, 2020, received for PCT Application PCT/EP2020/075417, Filed on Sep. 10, 2020, 12 pages. [cited by applicant]
Becker, “Fusion of Data From the Object-detecting Sensors of an Autonomous Vehicle”,Intelligent Transportation Systems, IEEE, Oct. 5-8, 1999, pp. 362-367. [cited by applicant]
Aeberhard, “Object-Level Fusion for Surround Environment Perception in Automated Driving Applications”, PhD thesis, XP055663290, May 31, 2017, 215 pages. [cited by applicant]
Omar et al., “Multiple Sensor Fusion for Detection, Classification and Tracking of Moving Objects in Driving Environments”,Thesis, XP055643862, Sep. 25, 2014, pp. 1-176. [cited by applicant]
Lee Hyungtae et al: “Dynamic belief fusion for object detection”, 2016 IEEE Winter Conference on Applications of Computer Vision (WACV), IEEE, Mar. 7, 2016 (Mar. 7, 2016), pp. 1-9, XP032904263,DOI: 10.1109/WACV.2016.747… [cited by applicant]
Razinkov Evgenii et al:“ALFA: Agglomerative Late Fusion Algorithm for Object Detection”, 2018 24th International Conference on Pattern Recognition (ICPR), IEEE, Aug. 20, 2018 (Aug. 20, 2018), pp. 2594-2599, XP033454012,… [cited by applicant]
Yong He, “A Novel Multi-source Vehicle Detection Algorithm based on Deep Learning”, 2018 14th IEEE International Conference on Signal Processing (ICSP), Aug. 12, 2018, p. 979-982. [cited by applicant]
Kiyosumi Kidono, “Pedestrian Recognition Integrating High-definition LIDAR with Vision”, Toyota Central R&D Labs., Inc. Toyohashi University of Technology Jun. 6, 2012, p. 207-1-207-6. [cited by applicant]