Apparatus and method for determining a loss function in a stacked display device thereof

1. A stacked display device comprising:

a display including a plurality of display panels; and

a processor processing a layer image for providing an image to the display,

wherein the display comprises:

a backlight irradiating a light,

a first panel, including a lens for diffusing the light irradiated from the backlight, and for displaying a first image of a first single color,

a second panel stacked on the first panel, and including a color filter for displaying a color image, and

a third panel stacked on the second panel, and for displaying a third image of a second single color, and

wherein the processor is configured to:

acquire first light field (LF) images of different views,

input the first LF images into a factorization model for converting LF images into layer images, acquire a first layer image for displaying the first image on the first panel, a second layer image for displaying a second image on the second panel, and a third layer image for displaying the third image on the third panel,

acquire a loss function based on the first LF images,

train the factorization model based on the loss function, and

control the display to, while displaying the first layer image on the first panel, display the second layer image on the second panel, and display the third layer image on the third panel.

2. The stacked display device of claim 1 ,

wherein the second panel is a color panel including the color filter, and

wherein the first panel and the third panel are monochrome panels not including color filters.

3. The stacked display device of claim 1 ,

wherein the lens is a multi-array lens, and

wherein the image provided to a user is different according to the view of the user viewing the stacked display device by the multi-array lens.

4. The stacked display device of claim 1 , wherein the processor is further configured to:

compare second LF images restored based on the first layer image, the second layer image, and the third layer image with the first LF images to acquire the loss function.

5. The stacked display device of claim 4 ,

wherein the second LF images are a collection of a plurality of images corresponding to a plurality of views, and

wherein the processor is further configured to:

convert the first layer image to correspond to a first view and acquire a 1-1 view layer image, and

restore the image corresponding to the first view among the second LF images based on the 1-1 view layer image, the second layer image, and the third layer image.

6. The stacked display device of claim 1 , wherein the processor is further configured to:

acquire first light field (LF) images of different views; and

input the first LF images into the factorization model for converting the LF image into the layer image, and acquire the second layer image, the third layer image, and a plurality of first view layer images corresponding to a plurality of views.

7. The stacked display device of claim 6 , wherein the processor is further configured to:

restore second LF images based on the plurality of respective first view layer images corresponding to the plurality of views, the second layer image, and the third layer image;

compare the restored second LF images with the first LF images and acquire the loss function; and

train the factorization model based on the loss function.

8. The stacked display device of claim 1 ,

wherein the first layer image and the third layer image are images of the first single color and the second single color, and

wherein the second layer image is a color image.

9. A method of controlling a stacked display device, the method comprising:

acquiring first light field (LF) images of different views;

inputting the first LF images into a factorization model for converting an LF image into a layer image, and acquiring a first layer image, a second layer image, and a third layer image;

acquiring second LF images restored based on the first layer image, the second layer image, and the third layer image;

acquiring a loss function by comparing the first LF images with the second LF images; and

training the factorization model based on the loss function.

10. The method of claim 9 ,

wherein the stacked display device includes a first panel, a second panel, and a third panel, and

wherein the first layer image is an image for being displayed on the first panel, the second layer image is an image for being displayed on the second panel, and the third layer image is an image for being displayed on the third panel.

11. The method of claim 10 ,

wherein, on the first panel, a multi-array lens for diffusing a light is arranged, and

wherein the image provided to a user is different according to a view of the user viewing the stacked display device by the multi-array lens.

12. The method of claim 11 ,

wherein the second LF images are a collection of a plurality of images corresponding to a plurality of views, and

wherein the acquiring the second LF images comprises:

acquiring a 1-1 view layer image by converting the first layer image to correspond to a first view, and

restoring an LF image corresponding to the first view among the second LF images based on the second layer image, the third layer image, and the 1-1 view layer image.

13. The method of claim 12 , wherein the converting the first layer image to correspond to the first view comprises:

acquiring the 1-1 view layer image by enlarging a pixel area corresponding to the first view among the pixel areas of the first layer image as much as a pitch of a multi-array lens.