System and method for generating realistic augmented reality content
In one embodiment, the system captures an image using a camera. The image is associated with a user viewpoint. The system identifies a surface in the image using a machine learning model. The surface has associated properties meeting one or more criteria for rendering a three-dimensional virtual space. The system determines relative positions and orientations of three-dimensional display elements to the surface. The system determines the three-dimensional virtual space based at least on the properties of the surface, the user viewpoint, and the relative positions and orientations of the three-dimensional display elements to the surface. The three-dimensional virtual space comprises the three-dimensional display elements, which are positioned behind the surface. The system renders the three-dimensional virtual space on the surface. The three-dimensional virtual space is visible through a display area on the surface as seen from the user viewpoint.
1. A method comprising, by a computing system:
capturing an image using a camera, wherein the image is associated with a user viewpoint;
identifying a vertical surface in the image using a machine learning model, wherein one or more properties associated with the vertical surface meet one or more criteria for rendering a three-dimensional virtual space corresponding to a virtual store;
determining relative positions and orientations of one or more three-dimensional display elements to the vertical surface, wherein the one or more three-dimensional display elements are positioned behind the vertical surface;
determining the three-dimensional virtual space based at least on the one or more properties of the vertical surface, the user viewpoint, and the relative positions and orientations of the three-dimensional display elements to the vertical surface, wherein the three-dimensional virtual space comprises the one or more three-dimensional display elements, and wherein the three-dimensional virtual space appears behind the vertical surface as seen from the user viewpoint;
rendering the three-dimensional virtual space for the vertical surface; and
displaying the three-dimensional virtual space, wherein the displayed three-dimensional virtual space, as seen from the user viewpoint, is visible within an area on the vertical surface and appears to replace contents within the area on the vertical surface.
2. The method of claim 1 , wherein the three-dimensional virtual space is rendered using one or more internal virtual light sources.
3. The method of claim 1 , wherein the associated properties of the vertical surface comprise one or more of:
a location;
a size;
a texture;
a material type;
a shape;
a flatness metric;
a lighting condition;
a direction;
a relative position to an object;
an angle in a three-dimensional space;
an associated object;
an associated person;
an associated place;
a blank space; or
a display content.
4. The method of claim 1 , wherein the one or more criteria are based on one or more thresholds for the one or more associated properties of the vertical surface.
5. The method of claim 1 , wherein the one or more three-dimensional display elements are related to one or more products associated with the virtual store.
6. The method of claim 1 , wherein the three-dimensional virtual space further comprises one or more two-dimensional display elements.
7. The method of claim 1 , wherein the three-dimensional virtual space comprises one or more of:
a model;
a product;
a brand name;
a text description;
a logo;
an illustration;
an artistical content;
a dynamical visual effect;
an interactive icon; or
a multi-media content.
8. The method of claim 1 , further comprising:
determining a user information associated with a user; and
determining the three-dimensional virtual space based on the user information associated with the user.
9. The method of claim 8 , wherein the user information associated with the user comprises one or more of:
a social network profile;
a search history;
a browsing history;
a purchase history;
an interest;
a preference;
a user setting;
an emotion state;
a behavioral state of the user;
a time related to the behavioral state of the user;
an interaction of the user with an object;
an indication of the user;
a user input;
a gesture; or
a command.
10. The method of claim 1 , wherein the vertical surface has one or more first display contents that are not overwritten by augmented reality contents, and wherein the method further comprises:
rendering one or more second display contents on the vertical surface in addition to the one or more first display contents of the vertical surface.
11. The method of claim 1 , further comprising:
measuring a status of a user when the user is near the three-dimensional virtual space rendered on the vertical surface; and
calculating one or more impression metrics based on the measured status of the user.
12. The method of claim 11 , wherein the status of the user comprises one or more of:
an emotion state;
a behavioral state of the user;
a time related to the behavioral state of the user;
an interaction of the user with an object;
an indication from the user;
a user input;
a setting choice of the user;
a gesture;
a vergence distance; or
a command.
13. The method of claim 11 , wherein the impression metrics comprise one or more of:
a time;
a location;
a gazing duration;
a gazing direction;
a gazing sequence;
an attention metric for an element;
a time interval since last impression; or
a reaction metric of the user.
14. The method of claim 11 , further comprising:
tracking an activity of a user after the user has an impression of the three-dimensional virtual space; and
determining a correlation between the activity of the user and the impression of the user on the three-dimensional virtual space.
15. The method of claim 1 , further comprising:
receiving from a user an interaction with the three-dimensional virtual space; and
adding an item related to the three-dimensional virtual space to a list based on the received interaction of the user.
16. The method of claim 15 , wherein the interaction comprises one or more of:
a selection on an interactive icon;
a gesture;
a voice command;
a vergence movement; or
a user input.
17. The method of claim 1 , wherein a bottom edge of the display area has a distance to a ground surface, wherein the distance is greater than a threshold distance for preventing a viewer from trying to walk into the three-dimensional virtual space.
18. The method of claim 1 , wherein the three-dimensional virtual space comprises one or more virtual walls, wherein the one or more three-dimensional display elements cast shadows on the one or more virtual walls of the three-dimensional virtual space, and wherein shapes of the casted shadows of the one or more three-dimensional display elements are dependent on relative positions of the one or more virtual walls in the three-dimensional virtual space.
19. One or more non-transitory computer-readable storage media embodying software that is operable when executed by a computing system to:
capture an image using a camera, wherein the image is associated with a user viewpoint;
identify a vertical surface in the image using a machine learning model, wherein one or more properties associated with the vertical surface meet one or more criteria for rendering a three-dimensional virtual space corresponding to a virtual store;
determine relative positions and orientations of one or more three-dimensional display elements to the vertical surface, wherein the one or more three-dimensional display elements are positioned behind the vertical surface;
determine the three-dimensional virtual space based at least on the one or more properties of the vertical surface, the user viewpoint, and the relative positions and orientations of the three-dimensional display elements to the vertical surface, wherein the three-dimensional virtual space comprises the one or more three-dimensional display elements, and wherein the three-dimensional virtual space appears behind the vertical surface as seen from the user viewpoint;
render the three-dimensional virtual space for the vertical surface; and
display the three-dimensional virtual space, wherein the displayed three-dimensional virtual space, as seen from the user viewpoint, is visible within an area on the vertical surface and appears to replace contents within the area on the vertical surface.
20. A system comprising:
one or more non-transitory computer-readable storage media embodying instructions;
one or more processors coupled to the storage media and operable to execute the instructions to:
capture an image using a camera, wherein the image is associated with a user viewpoint;
identify a vertical surface in the image using a machine learning model, wherein one or more properties associated with the vertical surface meet one or more criteria for rendering a three-dimensional virtual space corresponding to a virtual store;
determine relative positions and orientations of one or more three-dimensional display elements to the vertical surface, wherein the one or more three-dimensional display elements are positioned behind the vertical surface;
determine the three-dimensional virtual space based at least on the one or more properties of the vertical surface, the user viewpoint, and the relative positions and orientations of the three-dimensional display elements to the vertical surface, wherein the three-dimensional virtual space comprises the one or more three-dimensional display elements, and wherein the three-dimensional virtual space appears behind the vertical surface as seen from the user viewpoint;
render the three-dimensional virtual space for the vertical surface; and
display the three-dimensional virtual space, wherein the displayed three-dimensional virtual space, as seen from the user viewpoint, is visible within an area on the vertical surface and appears to replace contents within the area on the vertical surface.