Texture based auto exposure control algorithm
In one embodiment, a method includes accessing texture information generated and used by an autofocus unit to determine an autofocus configuration for a camera, the texture information including texture statistics detected in an environment sensed by the camera. The method further includes selecting a region of interest within a field of view of the camera based on the texture information, and determining an autoexposure configuration for the camera based on the region of interest. The method further includes capturing an image of the environment using the autoexposure configuration and the autofocus configuration, and tracking objects in the environment using the image.
1. A method comprising,
accessing texture information generated and used by an autofocus unit to determine an autofocus configuration for a camera, the texture information including texture statistics associated with a plurality of regions detected in an environment sensed by the camera, wherein the texture statistics comprise a plurality of texture values associating each region in the plurality of regions with an amount of texture present in the region;
selecting, based on the texture statistics, a region of interest comprising one or more regions within a field of view of the camera;
determining an autoexposure configuration for the camera based on the region of interest and the environment sensed by the camera;
capturing, by the camera, an image of the environment using the autoexposure configuration and the autofocus configuration; and
tracking objects within the environment using the image.
2. The method of claim 1 , wherein the camera is on an artificial reality device worn by a user.
3. The method of claim 1 , wherein the image captured by the camera comprises a first frame in a sequence of frames captured by the camera.
4. The method of claim 1 , further comprising:
determining autoexposure information including autoexposure statistics for the region of interest; and
using the autoexposure information in the determination of the autoexposure configuration.
5. The method of claim 1 , wherein the texture statistics are determined based on one or more of an infinite impulse response (IIR) filter and a finite impulse response (FIR) filter.
6. The method of claim 1 , wherein selecting the region of interest based on the texture statistics comprises:
assigning, for each of the plurality of regions within the field of view of the camera, a corresponding texture value based on the texture statistics;
comparing, for each of the plurality of regions, the corresponding texture value to a predetermined texture threshold; and
determining, for each of the plurality of regions, whether the corresponding texture value exceeds the predetermined texture threshold.
7. The method of claim 6 , wherein selecting the region of interest based on the texture statistics further comprises:
selecting one or more of the regions of the plurality of regions having a texture value which exceeds the predetermined texture threshold; and
discarding texture information of one or more of the regions of the plurality of regions having a texture value which is lower than the predetermined texture threshold.
8. The method of claim 6 , wherein the autoexposure configuration is determined by an automatic exposure control (AEC) algorithm using as an input the texture statistics of the one or more regions comprising the region of interest.
9. The method of claim 1 , further comprising:
receiving a feature point associated with a trackable object in the environment;
identifying, on an image captured by the camera, a location of the feature point;
identifying a region of the image containing the location;
selecting a portion of the field of view of the camera comprising the region containing the location; and
determining the autoexposure configuration based on the region of interest.
10. One or more computer-readable non-transitory non-volatile storage media embodying software that is operable when executed to:
access texture information generated and used by an autofocus unit to determine an autofocus configuration for a camera, the texture information including texture statistics associated with a plurality of regions detected in an environment sensed by the camera, wherein the texture statistics comprise a plurality of texture values associating each region in the plurality of regions with an amount of texture present in the region;
select, based on the texture statistics, a region of interest comprising one or more regions within a field of view of the camera;
determine an autoexposure configuration for the camera based on the region of interest and the environment sensed by the camera;
capture, by the camera, an image of the environment using the autoexposure configuration and the autofocus configuration; and
track objects within the environment using the image.
11. The computer-readable non-transitory non-volatile storage media of claim 10 , wherein the camera is on an artificial reality device worn by a user.
12. The computer-readable non-transitory non-volatile storage media of claim 10 , wherein the image captured by the camera comprises a first frame in a sequence of frames captured by the camera.
13. The computer-readable non-transitory non-volatile storage media of claim 10 , wherein the software is further operable when executed to:
determine autoexposure information including autoexposure statistics for the region of interest; and
use the autoexposure information in the determination of the autoexposure configuration.
14. The computer-readable non-transitory non-volatile storage media of claim 10 , wherein selecting the region of interest based on the texture statistics comprises:
assigning, for each of the plurality of regions within the field of view of the camera, a corresponding texture value based on the texture statistics;
comparing, for each of the plurality of regions, the corresponding texture value to a predetermined texture threshold; and
determining, for each of the plurality of regions, whether the corresponding texture value exceeds the predetermined texture threshold.
15. The computer-readable non-transitory non-volatile storage media of claim 14 , wherein selecting the region of interest based on the texture statistics further comprises:
selecting one or more of the regions of the plurality of regions having a texture value which exceeds the predetermined texture threshold; and
discarding texture information of one or more of the regions of the plurality of regions having a texture value which is lower than the predetermined texture threshold.
16. The computer-readable non-transitory non-volatile storage media of claim 14 , wherein the autoexposure configuration is determined by an automatic exposure control (AEC) algorithm using as an input the texture statistics of the one or more regions comprising the region of interest.
17. The computer-readable non-transitory non-volatile storage media of claim 10 , wherein the software is further operable when executed to:
receive a feature point associated with a trackable object in the environment;
identify, on an image captured by the camera, a location of the feature point;
identify a region of the image containing the location;
select a portion of the field of view of the camera comprising the region containing the location; and
determine the autoexposure configuration based on the region of interest.
18. The computer-readable non-transitory non-volatile storage media of claim 10 , wherein the texture statistics are determined based on one or more of an infinite impulse response (IIR) filter and a finite impulse response (FIR) filter.
19. A system comprising:
one or more processors; and
one or more computer-readable non-transitory non-volatile storage media coupled to one or more of the processors and comprising instructions operable when executed by one or more of the processors to cause the system to:
access texture information generated and used by an autofocus unit to determine an autofocus configuration for a camera, the texture information including texture statistics associated with a plurality of regions detected in an environment sensed by the camera, wherein the texture statistics comprise a plurality of texture values associating each region in the plurality of regions with an amount of texture present in the region;
select, based on the texture statistics, a region of interest comprising one or more regions within a field of view of the camera;
determine an autoexposure configuration for the camera based on the region of interest and the environment sensed by the camera;
capture, by the camera, an image of the environment using the autoexposure configuration and the autofocus configuration; and
track objects within the environment using the image.
20. The system of claim 19 , wherein the texture statistics are determined based on one or more of an infinite impulse response (IIR) filter and a finite impulse response (FIR) filter.