Object detection and tracking for autonomous driving utilizing shadows and reflections

Patent metadata

Field	Value
Publication number	US-12033333-B2
Application number	US-201816973845-A
Country	US
Kind code	B2
Filing date	Sep 5, 2018
Priority date	Sep 5, 2018
Publication date	Jul 9, 2024
Grant date	Jul 9, 2024

Abstract

An embodiment of a semiconductor package apparatus may include technology to analyze an electronic image to determine indirect information including one or more of shadow information and reflection information, and provide the indirect information to a vehicle guidance system. Other embodiments are disclosed and claimed.

First claim

We claim:

1. An electronic processing system, comprising:
a processor;
memory communicatively coupled to the processor; and
logic communicatively coupled to the processor to:
analyze an electronic image to determine indirect information including shadow information and reflection information,
provide the indirect information to a vehicle guidance system of an autonomous vehicle,
detect an object based on the indirect information by creating a shadow depth map containing a distance of the object from a light source and determining, for every pixel, when a point on a ground is shaded by projecting a depth value from the shadow depth map back onto the ground,
add the detected object to a scene,
make an autonomous vehicle guidance decision based on the scene to which the detected object is added, and
track the detected object based on the indirect information.

2. The electronic processing system of claim 1 , wherein the logic is further to:
make a vehicle guidance decision based at least in part on the indirect information.

3. The electronic processing system of claim 1 , wherein the processor comprises a graphics processor.

4. The electronic processing system of claim 1 , wherein tracking the detected object comprises adding the detected object to a navigation map.

5. The electronic processing system of claim 1 , wherein the logic is further to:
detect, over time, the detected object,
determine whether the detected object is moving, and
estimate, responsive to a determination that the detected object is moving the speed and trajectory of the detected object based on a history of prior movements of the detected object.

6. A semiconductor package apparatus, comprising: one or more substrates; and
logic coupled to the one or more substrates, wherein the logic is at least partly implemented in one or more of configurable logic and fixed-functionality hardware logic, the logic coupled to the one or more substrates to:
analyze an electronic image to determine indirect information including shadow information and reflection information,
provide the indirect information to a vehicle guidance system of an autonomous vehicle,
detect an object based on the indirect information by creating a shadow depth map containing a distance of the object from a light source and determining, for every pixel, when a point on a ground is shaded by projecting a depth value from the shadow depth map back onto the ground,
add the detected object to a scene,
make an autonomous vehicle guidance decision based on the scene to which the detected object is added, and
track the detected object based on the indirect information.

7. The semiconductor package apparatus of claim 6 , wherein the logic is further to:
make a vehicle guidance decision based at least in part on the indirect information.

8. The semiconductor package apparatus of claim 6 , wherein the logic coupled to the one or more substrates includes transistor channel regions that are positioned within the one or more substrates.

9. The semiconductor package apparatus of claim 6 , wherein tracking the detected object comprises adding the detected object to a navigation map.

10. The semiconductor package apparatus of claim 6 , wherein the logic is further to:
detect, over time, the detected object,
determine whether the detected object is moving, and
estimate, responsive to a determination that the detected object is moving the speed and trajectory of the detected object based on a history of prior movements of the detected object.

11. A method, comprising:
analyzing an electronic image to determine indirect information including shadow information and reflection information;
providing the indirect information to a vehicle guidance system of an autonomous vehicle,
detecting an object based on the indirect information by creating a shadow depth map containing a distance of the object from a light source and determining, for every pixel, when a point on a ground is shaded by projecting a depth value from the shadow depth map back onto the ground,
adding the detected object to a scene,
making an autonomous vehicle guidance decision based on the scene to which the detected object is added, and
tracking the detected object based on the indirect information.

12. The method of claim 11 , further comprising:
making a vehicle guidance decision based at least in part on the indirect information.

13. The method of claim 11 , further comprising: analyzing the electronic image with a graphics processor.

14. The method of claim 11 , wherein tracking the detected object comprises adding the detected object to a navigation map.

15. The method of claim 11 , further comprising:
detecting, over time, the detected object,
determining whether the detected object is moving, and
estimating, responsive to a determination that the detected object is moving the speed and trajectory of the detected object based on a history of prior movements of the detected object.

16. At least one non-transitory computer readable storage medium, comprising a set of instructions, which when executed by a computing device, cause the computing device to:
analyze an electronic image to determine indirect information including shadow information and reflection information;
provide the indirect information to a vehicle guidance system of an autonomous vehicle,
detect an object based on the indirect information by creating a shadow depth map containing a distance of the object from a light source and determining, for every pixel, when a point on a ground is shaded by projecting a depth value from the shadow depth map back onto the ground,
add the detected object to a scene,
make an autonomous vehicle guidance decision based on the scene to which the detected object is added, and
track the detected object based on the indirect information.

17. The at least one non-transitory computer readable storage medium of claim 16 , comprising a further set of instructions, which when executed by the computing device, cause the computing device to: make a vehicle guidance decision based at least in part on the indirect information.

18. The at least one non-transitory computer readable storage medium of claim 16 , comprising a further set of instructions, which when executed by the computing device, cause the computing device to: analyze the electronic image with a graphics processor.

19. The at least one non-transitory computer readable storage medium of claim 16 , wherein tracking the detected object comprises adding the detected object to a navigation map.

20. The at least one non-transitory computer readable storage medium of claim 16 , comprising a further set of instructions, which when executed by the computing device, cause the computing device to:
detect, over time, the detected object,
determine whether the detected object is moving, and
estimate, responsive to a determination that the detected object is moving the speed and trajectory of the detected object based on a history of prior movements of the detected object.

Assignees

• Intel Corp

Inventors

• Yang Wenlong
• Pohl Daniel
• Rider Tomer

Classifications

• G06T2207/30261

  Obstacle · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title

• G06T7/507

  from shading (G06T7/586 takes precedence) · CPC title

• G06T7/246 Primary

  using feature-based methods, e.g. the tracking of corners or segments · CPC title