Method and device for establishing the frontier between objects of a scene in a depth map

1. A method of rendering pixels of an image of a scene comprising a first object and at least a second object, a depth map being associated with the scene according to a point of view, at least a part of the first object and at least a part of the at least a second object being visible from the point of view associated with the depth map through a set of elements of the depth map, the first object and the at least a second object being modelled with surface elements, the method comprising:

from depth information associated with a point of the scene visible through a first element of said set of elements of the depth map, from depth information associated with said first element and from depth information associated with the other elements, called second elements, of said set of elements,

establishing a line associated with a direction associated with the frontier between the first object and the at least a second object, said line having as origin a corner of said set of elements located at an opposite side of a shadow generated by said first object on the depth map;

projecting orthogonally selected vertices of said surface elements visible through said set of elements onto said line, the selected vertices corresponding to the vertices having a normal vector oriented toward said line;

establishing a frontier between the first object and the at least a second object according to information representative of locations of the projected selected vertices onto said line and distances between the locations of the projected selected vertices and corresponding selected vertices; and

rendering pixels of said image of the scene associated with said depth map, the rendering of the pixels representing the first object and the at least second object being performed according to the established frontier.

2. The method according to claim 1 , wherein the set of elements correspond to a block of elements centred on said first element, the second elements surrounding the first element.

3. The method according to claim 1 , further comprising establishing whether said point belongs to the first object by comparing the position of a point resulting from the projection of said point into said first element with respect to said frontier.

4. The method according to claim 1 , wherein the depth information associated with the first element and each second element corresponds to the depth between the point of view and a fragment of the scene associated with respectively the first element and each second element, said fragment corresponding to the fragment visible from the point of view along viewing directions passing through the first element and each second element respectively.

5. The method according to claim 1 , further comprising storing, in each element of said set of elements, the coordinates of at least the selected vertices of the surface element visible through said element.

6. The method according to claim 1 , further comprising storing, in each element of said set of elements, an information representative of a variation of depth in said element in the space of the depth map.

7. Computer program product, characterized in that it comprises instructions of program code for executing steps of the method according to claim 1 , when said program is executed on a computer.

8. The method according to claim 1 , further comprising comparing, for said point of the scene visible through a first element of said set of elements of the depth map, the depth information associated with said point with the depth information associated with said first element and with the depth information associated with the second elements, the line being established from the comparison result.

9. A non-transitory processor readable medium having stored therein instructions for causing a processor to perform the method according to claim 1 .

10. A device configured for rendering pixels of an image of a scene comprising a first object and at least a second object, a depth map being associated with the scene according to a point of view, at least a part of the first object and at least a part of the at least a second object being visible from the point of view associated with the depth map through a set of elements of the depth map, the first object and the at least a second object being modelled with surface elements,

the device comprising at least one processor configured for;

from depth information associated with a point of the scene visible through a first element-of said set of elements of the depth map from depth information associated with said first element and from depth information associated with the other elements, called second elements, of said set of elements,

establishing a line associated with a direction associated with the frontier between the first object and the at least a second object, said line having as origin a corner of said set of elements located at an opposite side of a shadow generated by said first object on the depth map;

projecting orthogonally selected vertices of said surface elements visible through said set of elements onto said line, the selected vertices corresponding to the vertices having a normal vector oriented toward said line;

establishing a frontier between the first object and the at least a second object according to information representative of locations of the projected selected vertices onto said line and distances between the locations of the projected selected vertices and corresponding selected vertices; and

rendering pixels of said image of the scene associated with said depth map, the rendering of the pixels representing the first object and the at least second object being performed according to the established frontier.

11. The device according to claim 10 , wherein the at least one processor is a Graphical Processor Unit (GPU).

12. The device according to claim 10 , wherein the at least one processor is further configured for establishing whether said point belongs to the first object by comparing the position of a point resulting from the projection of said point into said first element with respect to said frontier.

13. The device according to claim 10 , wherein the at least one processor is further configured for storing, in each element of said set of elements, the coordinates of at least the selected vertices of the surface element visible through said element.

14. The device according to claim 10 , wherein the set of elements correspond to a block of elements centred on said first element, the second elements surrounding the first element.

15. The device according to claim 10 , wherein the depth information associated with the first element and each second element corresponds to the depth between the point of view and a fragment of the scene associated with respectively the first element and each second element, said fragment corresponding to the fragment visible from the point of view along viewing directions passing through the first element and each second element respectively.

16. The device according to claim 10 , wherein the at least one processor is further configured for storing, in each element of said set of elements, an information representative of a variation of depth in said element in the space of the depth map.

17. The device according to claim 10 , the at least one processor is further configured for comparing, for said point of the scene visible through a first element of said set of elements of the depth map, the depth information associated with said point with the depth information associated with said first element and with the depth information associated with the second elements, the line being established from the comparison result.