Systems and methods for selective enhancement of objects in images

1. A method for selective enhancement of an object in an image, the method comprising:

receiving as input at an interface an original image;

processing the original image using a neural network to detect one or more objects in the original image, wherein the original image includes a foreground and a background, and wherein the one or more objects include the background of the original image;

generating a neural network mask of the original image for the one or more objects in the original image;

applying one or more enhancements to the objects associated with the neural network mask and generating an enhanced image including the one or more enhancements to the objects, wherein applying the one or more enhancements includes:

generating a refined mask for each pixel of at least the original image associated with the background; and

generating a dilated mask including a dilation or indentation from a border extending between the foreground and the background; and

generating a combined image, the combined image including the original image combined with the one or more enhancements to the objects of the enhanced image.

2. The method of claim 1 , wherein the one or more objects include a sky in the original image.

3. The method of claim 2 , further comprising determining a category type of the sky in the original image.

4. The method of claim 3 , wherein the category type includes at least one of a blue day sky, a sunset sky, a dawn sky, a cloudy sky, a stormy sky, a night sky, or a foggy sky.

5. The method of claim 1 , further comprising generating a blurred mask, the blurred mask including a blur along a border extending between the foreground and the background.

6. The method of claim 1 , wherein the one or more enhancements applied to the objects associated with the neural network mask are applied to the background independently from the foreground of the original image.

7. The method of claim 1 , wherein the one or more enhancements applied to the objects associated with the neural network mask are applied to the background without affecting the foreground of the original image.

8. The method of claim 1 , further comprising generating a saturation corrected image including saturation enhancement of the original image.

9. The method of claim 1 , further comprising generating a color improved image including color enhancement of the original image.

10. The method of claim 1 , further comprising generating a reduced noise image by reducing noise associated with a night sky in the original image.

11. A non-transitory computer-readable medium storing instructions at least for selective enhancement of an object in an image that are executable by a processing device, wherein execution of the instructions by the processing device causes the processing device to:

receive as input at an interface an original image;

process the original image using a neural network to detect one or more objects in the original image, wherein the original image includes a foreground and a background, and wherein the one or more objects include the background of the original image;

generate a neural network mask of the original image for the one or more objects in the original image;

apply one or more enhancements to the objects associated with the neural network mask and generating an enhanced image including the one or more enhancements to the objects, wherein applying the one or more enhancements includes:

generating a refined mask for each pixel of at least the original image associated with the background; and

generating a dilated mask including a dilation or indentation from a border extending between the foreground and the background; and

generate a combined image, the combined image including the original image combined with the one or more enhancements to the objects of the enhanced image.

12. The non-transitory computer-readable medium of claim 11 , wherein the one or more objects include a sky in the original image.

13. The non-transitory computer-readable medium of claim 12 , wherein the processing device determines a category type of the sky in the original image.

14. The non-transitory computer-readable medium of claim 13 , wherein the category type includes at least one of a blue day sky, a sunset sky, a dawn sky, a cloudy sky, a stormy sky, a night sky, or a foggy sky.

15. The non-transitory computer-readable medium of claim 11 , wherein the processing device generates a blurred mask, the blurred mask including a blur along a border extending between the foreground and the background.

16. The non-transitory computer-readable medium of claim 11 , wherein the one or more enhancements applied to the objects associated with the neural network mask are applied to the background independently from the foreground of the original image.

17. The non-transitory computer-readable medium of claim 11 , wherein the one or more enhancements applied to the objects associated with the neural network mask are applied to the background without affecting the foreground of the original image.

18. The non-transitory computer-readable medium of claim 11 , wherein the processing device generates a saturation corrected image including saturation enhancement of the original image.

19. The non-transitory computer-readable medium of claim 11 , wherein the processing device generates a color improved image including color enhancement of the original image.

20. The non-transitory computer-readable medium of claim 11 , wherein the processing device generates a reduced noise image by reducing noise associated with a night sky in the original image.