Depth-based projection of image-based content

1. A projection system, comprising:

circuitry communicatively coupled to an image projection sub-system that comprises illumination circuitry, wherein the circuitry is configured to:

receive a sequence of image frames to be projected on a physical surface of a three-dimensional (3D) structure;

control a depth sensor to acquire depth data associated with the physical surface;

feed a first input comprising the acquired depth data and a first image frame of the received sequence of image frames to a neural network-based model;

receive a set of geometric correction values as a first output of the neural network-based model for the fed first input;

modify the first image frame based on the received set of geometric correction values; and

control the illumination circuitry to project the modified first image frame onto the physical surface.

2. The projection system according to claim 1 , wherein the circuitry is further configured to:

determine first depth values associated with corners of the physical surface;

generate virtual depth data for the physical surface based on the determined first depth values;

compare the acquired depth data with the generated virtual depth data to generate a comparison result;

apply a clustering method on the comparison result to divide the physical surface into surface clusters, each of which corresponds to one or more surface portions of the physical surface;

calculate a smoothness value for each of the surface clusters;

categorize one or more portions of the physical surface into different categories based on the calculated smoothness value; and

determine an even surface portion of the physical surface, based on the categorization,

wherein the modified first image frame is projected onto the determined even surface portion.

3. The projection system according to claim 1 , wherein the neural network-based model comprises a convolutional neural network, and the first input, at first, is fed to the convolutional neural network.

4. The projection system according to claim 3 , wherein the circuitry is further configured to:

input the acquired depth data to the convolutional neural network;

receive an output of the convolutional neural network for the input depth data,

wherein the received output comprises a clustering result which is indicative of one or more even surface portions of the physical surface and one or more uneven surface portions of the physical surface; and

determine an even surface portion of a maximum size from among that of the one or more even surface portions, wherein the even surface portion is determined based on the clustering result.

5. The projection system according to claim 4 , wherein the circuitry is further configured to:

extract coordinate values associated with corners of the determined even surface portion;

resize the first image frame based on the extracted coordinate values; and

modify the resized first image frame based on the received set of geometric correction values,

wherein the modified first image frame is projected onto the determined even surface portion of the physical surface.

6. The projection system according to claim 4 , wherein the modified first image frame is projected onto the determined even surface portion.

7. The projection system according to claim 4 , wherein the neural network-based model further includes an artificial neural network, an input layer of which is coupled to an output layer of the convolutional neural network.

8. The projection system according to claim 7 , wherein the circuitry is further configured to:

determine a set of projection length values associated with corners of the determined even surface portion; and

input the determined set of projection length values to the artificial neural network,

wherein the artificial neural network generates, based on the input set of projection length values, the first output that includes the set of geometric correction values.

9. The projection system according to claim 1 , wherein the modification of the first image frame includes an application of an image warping operation on the first image frame while maintaining an aspect ratio of the first image frame.

10. The projection system according to claim 1 , further comprising the image projection sub-system that comprises the illumination circuitry.

11. A method, comprising:

in a projection system communicatively coupled to an image projection sub-system:

receiving a sequence of image frames to be projected on a physical surface of a three-dimensional (3D) structure;

controlling a depth sensor to acquire depth data associated with the physical surface;

feeding a first input comprising the acquired depth data and a first image frame of the received sequence of image frames to a neural network-based model;

receiving a set of geometric correction values as a first output of the neural network-based model for the fed first input;

modifying the first image frame based on the received set of geometric correction values; and

controlling illumination circuitry of the image projection sub-system to project the modified first image frame onto the physical surface.

12. The method according to claim 11 , further comprising:

determining first depth values associated with corners of the physical surface;

generating virtual depth data for the physical surface based on the determined first depth values;

comparing the acquired depth data with the generated virtual depth data to generate a comparison result;

applying a clustering method on the comparison result to divide the physical surface into surface clusters, each of which corresponds to one or more surface portions of the physical surface;

calculating a smoothness value for each of the surface clusters;

categorizing one or more portions of the physical surface into different categories based on the calculated smoothness; and

determining an even surface portion of the physical surface, based on the categorization,

wherein the modified first image frame is projected onto the determined even surface portion.

13. The method according to claim 11 , wherein the neural network-based model comprises a convolutional neural network, and the first input, at first, is fed to the convolutional neural network.

14. The method according to claim 13 , further comprising:

inputting the acquired depth data to the convolutional neural network;

receiving an output of the convolutional neural network for the input depth data,

wherein the received output comprises a clustering result which is indicative of one or more even surface portions of the physical surface and one or more uneven surface portions of the physical surface; and

determining an even surface portion of a maximum size from among that of the one or more even surface portions, wherein the even surface portion is determined based on the clustering result.

15. The method according to claim 14 , wherein the modified first image frame is projected onto the determined even surface portion.

16. The method according to claim 14 , further comprising:

extracting coordinate values associated with corners of the determined even surface portion;

resizing the first image frame based on the extracted coordinate values.

modifying the resized first image frame based on the received set of geometric correction values; and

projecting the modified first image frame onto the determined even surface portion of the physical surface.

17. The method according to claim 14 , wherein the neural network-based model further includes an artificial neural network, an input layer of which is coupled to an output layer of the convolutional neural network.

18. The method according to claim 17 , further comprising:

determining a set of projection length values associated with corners of the determined even surface portion; and

inputting the determined set of projection length values to the artificial neural network,

wherein the artificial neural network generates, based on the input set of projection length values, the first output that includes the set of geometric correction values.

19. The method according to claim 11 , wherein the modification of the first image frame includes an application of an image warping operation on the first image frame while maintaining an aspect ratio of the first image frame.

20. A non-transitory computer-readable storage medium configured to store instructions that, in response to being executed, causes a projection system to perform operations, the operations comprising:

receiving a sequence of image frames to be projected on a physical surface of a three-dimensional (3D) structure;

controlling a depth sensor to acquire depth data associated with the physical surface;

feeding a first input comprising the acquired depth data and a first image frame of the received sequence of image frames to a neural network-based model;

receiving a set of geometric correction values as a first output of the neural network-based model for the fed first input;

modifying the first image frame based on the received set of geometric correction values; and

controlling illumination circuitry of an image projection sub-system communicatively coupled to the projection system, to project the modified first image frame onto the physical surface.