Techniques for determining settings for a content capture device

1. A method for computing a total weight array, the method comprising:

receiving an image frame captured by a content capture device;

identifying a plurality of objects in the image frame, wherein each object of the plurality of objects is represented by one of a plurality of pixel groups corresponding to a shape of each object of the plurality of objects;

providing a plurality of neural networks;

calculating, for each object of the plurality of objects, an object weight using a corresponding neural network of the plurality of neural networks; and

computing the total weight array by summing the object weight for each of the plurality of objects.

2. The method of claim 1 wherein each of the plurality of neural networks comprises a different neural network.

3. The method of claim 1 wherein each object of the plurality of objects is associated with a row r and a column c of the image frame.

4. The method of claim 3 wherein the total weight array is

w

T

[

r

,

c

]

=

∑

i

=

0

N

o

w

i

[

r

,

c

]

,

where N o is the number of objects and w i [r, c] is the object weight for each of the plurality of objects.

5. The method of claim 1 wherein each object weight comprises a single value.

6. The method of claim 5 wherein the single value is applied to all pixels in each of the pixel groups of the plurality of pixel groups.

7. The method of claim 1 further comprising:

identifying a target luma value for the image frame;

calculating an image luma value using the total weight array; and

computing a difference between the image luma value and the target luma value; and

updating a setting of the content capture device based upon the computed difference.

8. The method of claim 1 further comprising:

receiving, by each neural network of the plurality of neural networks, a plurality of inputs corresponding to the object associated with the corresponding neural network; and

outputting, by each neural network of the plurality of neural networks, a single weight for the object associated with the corresponding neural network.

9. The method of claim 1 further comprising receiving, by each neural network of the plurality of neural networks, a plurality of attributes as inputs to each neural network of the plurality of neural networks.

10. A method comprising:

receiving an image captured by a content capture device;

identifying a target luma value for the image;

providing a plurality of neural networks;

identifying a plurality of objects in the image, wherein each of the plurality of objects is represented by a pixel group corresponding to a shape of each object of the plurality of objects;

calculating, for each object of the plurality of objects, an object weight using a corresponding neural network of the plurality of neural networks;

defining a first set of pixel groups associated with the plurality of objects;

defining a second set of pixel groups not associated with the plurality of objects;

calculating a pixel group luma value for each pixel group of the first set of pixel groups;

multiplying the pixel group luma value by the object weight to provide a weighted pixel group luma value for each pixel group of the first set of pixel groups; and

calculating a total luma value for the image.

11. The method of claim 10 further comprising:

computing a difference between the total luma value and the target luma value; and

updating a setting of the content capture device based upon the computed difference.

12. The method of claim 10 wherein the image comprises one image of a stream of images.

13. The method of claim 10 wherein the image comprises pixels, each having a pixel luma value, and the target luma value corresponds to an average of the pixel luma values.

14. The method of claim 10 wherein the image comprises pixels, each having a pixel luma value and a weight, and the target luma value corresponds to a weighted average of the pixel luma values.

15. The method of claim 10 further comprising identifying one or more attributes for each of the plurality of objects in the image.

16. The method of claim 15 wherein the one or more attributes include at least one of a priority weight array for object priority, a size weight array for object size, a distance weight array for object distance, or a gaze weight array for eye gaze.

17. The method of claim 15 wherein each corresponding neural network uses the one or more attributes as input.

18. The method of claim 10 wherein the pixel group luma value comprises an average of luma values for each pixel of the pixel group.

19. The method of claim 10 wherein the total luma value equals a summation of the weighted pixel group luma value for each pixel group of the first set of pixel groups times the pixel group luma value for each pixel group of the first set of pixel groups.

20. The method of claim 10 wherein each of the plurality of neural networks comprises a different neural network.