ZNGA: PATENT#9216346,DATE,12,22,2015,3D,MORE,QUICKLY,MORE,EASILY,THAN,TRADITIONAL,METHODS...
"the ability to more quickly and more easily author 3D surface detail for real-time applications than is generally available using traditional methods;"
Systems and methods described herein may use a multi-image projection for encoding primary surface detail of 3D worlds in real-time applications. An algorithm for improved image projection camera path trajectories, sampling selection, and blending may be implemented by these systems and methods to facilitate smooth transitions during movements. In addition, a back-projection technique may be implemented by the systems and methods to propagate changes to all projections for manual modifications of 3D surface detail via single projection image modifications.
The systems and methods discussed herein may provide one or more benefits, including for example:
the ability to more quickly and more easily author 3D surface detail for real-time applications than is generally available using traditional methods;
improved layouts for 3D world content that improve streaming behavior by drastically lowering the size of initial content downloaded by the end user
...
What is claimed is:
1. A method comprising: in response to receiving a request from a client device to initiate execution of a game, generating, by a server comprising one or more processors, a collection of images, each image of the collection of images depicting at least a portion of a three-dimensional (3D) space from one of multiple projection nodes along a projection path, the projection path connecting the multiple projection nodes within the 3D space; providing, by the server, a first portion of the collection of images to the client device based on a positioning of a first projection node relative to a position of a player that represents a user within the 3D space, the client device being configured to render at least a portion of the 3D space from the first portion of the collection of images; and providing, by the server, a second portion of the collection of images to the client device based on movement of the player within the 3D space, the client device being configured to render at least a further portion of the 3D space from the second portion of the collection of images...
...Image projection may encompass all surface texture detail of any one or more objects visible from a given camera. For given known camera positions and directions, the systems and methods may pre-generate one or more projections and may stream in (e.g., download or retrieve) any one or more additional projections that may be used to span the space to account for movement of an in-world (e.g., in-game) camera in a similar direction and location. This may ensure that the surfaces in view from the in-world camera are filled by the projection data. UV mapping may hence be omitted entirety, since the projection itself represents the UV coordinates for a given camera transformation. Thus, quality of data may be increased, since the projection provides exactly what was intended from that given projection camera view, with minimal stretching or warping in UV-space. In addition, the data footprint (e.g., in a memory) may be reduced, since a relatively low number (e.g., three) of projection images may suffice to span an entire space at any single given location of the player from a given camera view. Additionally, the scene may be easier to author and modify, in the absence of UV maps to generate or limitations of surface detail on runtime hardware...
Date of Patent: December 22, 2015
Assignee: Zynga Inc.
Source
ZYNGA'S,PATENT#9216346
