IP Library Granted Patent US 12,430,814
Granted Patent B2
US 12,430,814 · App. 18/474,189 · Granted Sep 30, 2025

Image generation method and apparatus, and storage medium

Inventors: Guangliang Cheng (Hong Kong, CN); Jianping Shi (Hong Kong, CN); Yuji Yasui (Tokyo, JP); Hideki Matsunaga (Tokyo, JP); Kaname Tomite (Tokyo, JP)
Assignees: SENSETIME GROUP LIMITED; HONDA MOTOR CO., LTD.
G06T11/00G06T5/40G06T5/77G06T2207/20081G06T2207/20084
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 12,430,814
App. No.
18/474,189
Granted
Sep 30, 2025
Kind
B2
Abstract

An image generation method includes: an original image is converted into an image to be processed under a specific scenario; in the image to be processed, a region to be adjusted of which image rendering state does not match scenario information of the specific scenario is determined; the image rendering state of the region to be adjusted is related to light illuminated on a target object in the original image, and the target object includes a vehicle; a target rendering state in the region to be adjusted is determined; the target rendering state matches the scenario information; and in the image to be processed, the image rendering state of the region to be adjusted is adjusted to the target rendering state, to obtain a target image.

Claims (73)

1. An image generation method, performed by an electronic device, the method comprising:

converting an original image into an image to be processed under a specific scenario;

determining, in the image to be processed, a region to be adjusted of which image rendering state does not match scenario information of the specific scenario, wherein the image rendering state of the region to be adjusted is related to light illuminated on a target object in the original image, and the target object comprises a vehicle;

determining a target rendering state in the region to be adjusted, wherein the target rendering state matches the scenario information; and

adjusting, in the image to be processed, the image rendering state of the region to be adjusted to the target rendering state, to obtain a target image,

wherein the image rendering state comprises a shadow rendering state; and the determining the target rendering state in the region to be adjusted comprises:

determining, in the image to be processed, a reference region of which shadow rendering state matches the scenario information; and

determining the shadow rendering state of the reference region as the target rendering state, and

wherein the determining, in the image to be processed, the region to be adjusted of which image rendering state does not match scenario information of the specific scenario comprises:

in a case where the scenario information indicates a night scenario without a light source or a daytime scenario without sunshine, determining a shadow region in the image to be processed as the region to be adjusted.

2. The method of claim 1 , wherein the determining, in the image to be processed, the region to be adjusted of which image rendering state does not match scenario information of the specific scenario comprises:

in a case where the scenario information indicates a night scenario with a light source or a daytime scenario with sunshine, determining, according to a direction in which light irradiates the target object, a shadow region capable of being generated by the target object; and

determining a shadow region other than a first shadow region within the shadow region in the image to be processed as the region to be adjusted, wherein the first shadow region is an intersection of the shadow region in the image to be processed and the shadow region capable of being generated by the target object.

3. The method of claim 1 , wherein the shadow rendering state of the reference region is a shadow-free state; and the adjusting, in the image to be processed, the image rendering state of the region to be adjusted to the target rendering state, to obtain the target image comprises:

adjusting, in the image to be processed, an image of the region to be adjusted to be in a shadow-free state according to an image of the reference region, to obtain the target image.

4. The method of claim 3 , wherein the adjusting, in the image to be processed, the image of the region to be adjusted to be in the shadow-free state according to the image of the reference region, to obtain the target image comprises:

adjusting a grayscale image of the region to be adjusted according to the image of the reference region, to obtain an adjusted grayscale map;

generating, based on the adjusted grayscale map, an image of a replacement region of which shadow rendering state is the shadow-free state; and

replacing, in the image to be processed, the image of the region to be adjusted with a generated image of the replacement region, to generate the target image.

5. The method of claim 4 , wherein the adjusting the grayscale image of the region to be adjusted according to the image of the reference region, to obtain the adjusted grayscale map comprises:

determining a histogram of the reference region according to the image of the reference region;

adjusting a histogram of the region to be adjusted based on the histogram of the reference region, to obtain an adjusted histogram; and

determining the adjusted grayscale map based on the adjusted histogram.

6. The method of claim 1 , wherein the image rendering state comprises a shadow rendering state, and the determining, in the image to be processed, the region to be adjusted of which image rendering state does not match the scenario information comprises:

in a case where the scenario information indicates a night scenario with a light source or a daytime scenario with sunshine, determining, according to a direction in which light irradiates the target object, a shadow region capable of being generated by the target object; and

determining the shadow region capable of being generated by the target object as the region to be adjusted.

7. The method of claim 6 , wherein the adjusting, in the image to be processed, the image rendering state of the region to be adjusted to the target rendering state, to obtain the target image comprises:

determining a light parameter in the specific scenario; and

determining, based on the light parameter, a light intensity of a surface of the target object that is able to generate the shadow region, and rendering the region to be adjusted to obtain the target image.

8. The method of claim 1 , wherein the original image is an image acquired without rain/snow, and the specific scenario is a rain/snow scenario, and

the adjusting, in the image to be processed, the image rendering state of the region to be adjusted to the target rendering state, to obtain the target image comprises:

adding raindrops/snowflakes in the original image, to obtain the target image.

9. An image generation apparatus, comprising:

a memory storing processor-executable instructions; and

a processor configured to execute the stored processor-executable instructions to perform operations of:

converting an original image into an image to be processed under a specific scenario;

determining, in the image to be processed, a region to be adjusted of which image rendering state does not match scenario information of the specific scenario, wherein the image rendering state of the region to be adjusted is related to light illuminated on a target object in the original image, and the target object comprises a vehicle;

determining a target rendering state in the region to be adjusted, wherein the target rendering state matches the scenario information; and

adjusting, in the image to be processed, the image rendering state of the region to be adjusted to the target rendering state, to obtain a target image,

wherein the image rendering state comprises a shadow rendering state; and the determining the target rendering state in the region to be adjusted comprises:

determining, in the image to be processed, a reference region of which shadow rendering state matches the scenario information; and

determining the shadow rendering state of the reference region as the target rendering state, and

wherein the determining, in the image to be processed, the region to be adjusted of which image rendering state does not match scenario information of the specific scenario comprises:

in a case where the scenario information indicates a night scenario without a light source or a daytime scenario without sunshine, determining a shadow region in the image to be processed as the region to be adjusted.

10. The apparatus of claim 9 , wherein the determining, in the image to be processed, the region to be adjusted of which image rendering state does not match scenario information of the specific scenario comprises:

in a case where the scenario information indicates a night scenario with a light source or a daytime scenario with sunshine, determining, according to a direction in which light irradiates the target object, a shadow region capable of being generated by the target object; and

determining a shadow region other than a first shadow region within the shadow region in the image to be processed as the region to be adjusted, wherein the first shadow region is an intersection of the shadow region in the image to be processed and the shadow region capable of being generated by the target object.

11. The apparatus of claim 9 , wherein the shadow rendering state of the reference region is a shadow-free state; and the adjusting, in the image to be processed, the image rendering state of the region to be adjusted to the target rendering state, to obtain the target image comprises:

adjusting, in the image to be processed, an image of the region to be adjusted to be in a shadow-free state according to an image of the reference region, to obtain the target image.

12. The apparatus of claim 11 , wherein the adjusting, in the image to be processed, the image of the region to be adjusted to be in the shadow-free state according to the image of the reference region, to obtain the target image comprises:

adjusting a grayscale image of the region to be adjusted according to the image of the reference region, to obtain an adjusted grayscale map;

generating, based on the adjusted grayscale map, an image of a replacement region of which shadow rendering state is the shadow-free state; and

replacing, in the image to be processed, the image of the region to be adjusted with a generated image of the replacement region, to generate the target image.

13. The apparatus of claim 12 , wherein the adjusting the grayscale image of the region to be adjusted according to the image of the reference region, to obtain the adjusted grayscale map comprises:

determining a histogram of the reference region according to the image of the reference region;

adjusting a histogram of the region to be adjusted based on the histogram of the reference region, to obtain an adjusted histogram; and

determining the adjusted grayscale map based on the adjusted histogram.

14. The apparatus of claim 9 , wherein the image rendering state comprises a shadow rendering state, and the determining, in the image to be processed, the region to be adjusted of which image rendering state does not match the scenario information comprises:

in a case where the scenario information indicates a night scenario with a light source or a daytime scenario with sunshine, determining, according to a direction in which light irradiates the target object, a shadow region capable of being generated by the target object; and

determining the shadow region capable of being generated by the target object as the region to be adjusted.

15. The apparatus of claim 14 , wherein the adjusting, in the image to be processed, the image rendering state of the region to be adjusted to the target rendering state, to obtain the target image comprises:

determining a light parameter in the specific scenario; and

determining, based on the light parameter, a light intensity of a surface of the target object that is able to generate the shadow region, and rendering the region to be adjusted to obtain the target image.

16. A non-transitory computer-readable storage medium having stored thereon computer-executable instructions that, when executed by a processor, cause the processor to implement operations of:

converting an original image into an image to be processed under a specific scenario;

determining, in the image to be processed, a region to be adjusted of which image rendering state does not match scenario information of the specific scenario, wherein the image rendering state of the region to be adjusted is related to light illuminated on a target object in the original image, and the target object comprises a vehicle;

determining a target rendering state in the region to be adjusted, wherein the target rendering state matches the scenario information; and

adjusting, in the image to be processed, the image rendering state of the region to be adjusted to the target rendering state, to obtain a target image,

wherein the image rendering state comprises a shadow rendering state, and the determining the target rendering state in the region to be adjusted comprises:

determining, in the image to be processed, a reference region of which shadow rendering state matches the scenario information; and

determining the shadow rendering state of the reference region as the target rendering state, and

wherein the determining, in the image to be processed, the region to be adjusted of which image rendering state does not match scenario information of the specific scenario comprises:

in a case where the scenario information indicates a night scenario without a light source or a daytime scenario without sunshine, determining a shadow region in the image to be processed as the region to be adjusted.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 25, 2023
From: CHENG, GUANGLIANG; SHI, JIANPING; YASUI, YUJI; MATSUNAGA, HIDEKI; TOMITE, KANAME
To: SENSETIME GROUP LIMITED; HONDA MOTOR CO., LTD.
Reel/Frame 065017/0994 →
Priority Claims (1)
CN 202110351943.9 · Mar 31, 2021 · national
Continuity (2)
Continuation PCTCN2022074059 · Jan 26, 2022
Related Publication 20240013453A1 · Jan 11, 2024
References Cited (33)
US 6873327B1 · Edwards · 2005 [cited by examiner]
US 7295716B1 · Chinen · 2007 [cited by examiner]
US 8406519B1 · Lim · 2013 [cited by examiner]
US 10885701B1 · Patel · 2021 [cited by applicant]
US 20200066030A1 · Naik et al. · 2020 [cited by applicant]
US 20200288067A1 · Watazawa · 2020 [cited by examiner]
US 20210118112A1 · Huang · 2021 [cited by applicant]
US 20210295571A1 · Sun · 2021 [cited by examiner]
US 20220327769A1 · Tsai · 2022 [cited by examiner]
US 20230088801A1 · Lawrence · 2023 [cited by examiner]
US 20230206568A1 · Du · 2023 [cited by examiner]
CN 102136132A · 2011 [cited by applicant]
CN 106327454A · 2017 [cited by applicant]
CN 109377502A · 2019 [cited by applicant]
CN 109887066A · 2019 [cited by applicant]
CN 110264413A · 2019 [cited by applicant]
CN 110490960A · 2019 [cited by applicant]
CN 111260769A · 2020 [cited by applicant]
CN 111583097A · 2020 [cited by applicant]
CN 111951157A · 2020 [cited by applicant]
CN 112419328A · 2021 [cited by applicant]
CN 112991158A · 2021 [cited by applicant]
JP 2010033321A · 2010 [cited by applicant]
JP 2015149611A · 2015 [cited by applicant]
JP 2016072694A · 2016 [cited by applicant]
JP 2017016663A · 2017 [cited by applicant]
JP 2017152866A · 2017 [cited by applicant]
WO 2016054800A1 · 2016 [cited by applicant]
WO 2020220807A1 · 2020 [cited by applicant]
International Search Report in the international application No. PCT/CN2022/074059, mailed on Apr. 1, 2022, 3 pages. [cited by applicant]
English translation of the Written Opinion of the International Search Authority in the international application No. PCT/CN2022/074059, mailed on Apr. 1, 2022, 6 pages. [cited by applicant]
Satoru Nakanishi et al, “Deep Photo Relighting: Changing lighting conditions for free viewpoint images using deep learning”, Research Report Audiovisual Complex Information Processing (AVM), Feb. 27, 2018, vol. 2018-AVM… [cited by applicant]
Yohei Ogura et al, “Illumination Estimation and Relighting based on Normal Map from an RGB-D Camera”, IPSJ SIG Technical Report, Jan. 15, 2015, vol. 2015-CVIM-195 No. 32, pp. 1-4. [cited by applicant]