Method for generating layered depth data of a scene
The invention relates to layered depth data. In multi-view images, there is a large amount of redundancy between images. Layered Depth Video format is a well-known formatting solution for formatting multi-view images which reduces the amount of redundant information between images. In LDV, a reference central image is selected and information brought by other images of the multi-view images that are mainly regions occluded in the central image are provided. However, LDV format contains a single horizontal occlusion layer, and thus fails rendering viewpoints that uncover multiple layers dis-occlusions. The invention uses light filed content which offers disparities in every directions and enables a change in viewpoint in a plurality of directions distinct from the viewing direction of the considered image enabling to render viewpoints that may uncover multiple layer dis-occlusions which may occurs with complex scenes viewed with wide inter-axial distance.
1. A computer implemented method:
receiving an image of a scene from a light filed content;
generating a disparity estimation to obtain a first and a second depth map of said scene, said image being viewed according to a given viewing direction;
generating in a first direction distinct from the viewing direction of the image from the light field content, a first set of occlusion information associated with the image from the light field content by using said first depth map;
generating, in at least a second direction distinct from the viewing direction of the image from the light field content and from the first direction, at least a second set of occlusion information associated with the image from the light field content by using said second depth map, wherein the first set of occlusion information and the second set of occlusion information are merged together in order to generate a third set of occlusion information by analyzing said first and second depth maps in areas of said image that are inconsistent;
wherein computing the first and the second set of occlusion information includes comparing the image from the light field content with another adjacent image from the light field content in the first direction and the second direction, respectively; and
generating a layered depth data of said scene by aggregating the image from the light field content, the first and second depth maps, and the third set of occlusion information.
2. An apparatus comprising a processor configured to:
receive an image of a scene from a light filed content;
generate a disparity estimation to obtain a first depth map and a second depth map of said scene, said image being viewed according to a given viewing direction;
generate, in a first direction distinct from the viewing direction of the image from the light field content, a first set of occlusion information associated with the image from the light field content by using said first depth map;
generate, in at least a second direction distinct from the viewing direction of the image from the light field content and from the first direction, at least a second set of occlusion information associated with the image from the light field content by using said second depth map, wherein the first set of occlusion information and the second set of occlusion information are merged together in order to generate a third set of occlusion information;
wherein computing the first and the second set of occlusion information includes comparing the image from the light field content with another adjacent image from the light field content in the first direction and the second direction, respectively; and
generate a layered depth data of said scene by aggregating the image from the light field content, the first and second depth maps, and the third set of occlusion information.
3. A method comprising:
receiving an image of a scene from a light filed content;
generating a disparity estimation to obtain a first and a second depth map of said scene, said image being viewed according to a given viewing direction;
generating in a first direction distinct from the viewing direction of the image from the light field content, a first set of occlusion information associated with the image from the light field content by using said first depth map;
generating, in at least a second direction distinct from the viewing direction of the image from the light field content and from the first direction, at least a second set of occlusion information associated with the image from the light field content by using said second depth map, wherein the first set of occlusion information and the second set of occlusion information are merged together in order to generate a third set of occlusion information by analyzing said first and second depth maps in areas of said image that are inconsistent;
wherein computing the first and the second set of occlusion information includes comparing the image from the light field content with another adjacent image from the light field content in the first direction and the second direction, respectively; and
generating a layered depth data of said scene by aggregating the image from the light field content, the first and second depth maps, and the third set of occlusion information.