- Number
- 9959654
- Published
- 2018-05-01
- Filed
- 2011-12-08
- Assignee
- Disney Enterprises, Inc.
- Inventors
- Lee; Gene S.
- CPC
- G06T13/40
- Verdict
- Low Notable software
- Source
- Google Patents · FreePatentsOnline
The keeper's note
Animation data selection for data-driven model.
Abstract
A set of animation data for an element in an animation is statistically sampled to obtain a common context. The common context is a subset of a plurality of frames of the set of animation data. Further, output of a data-driven model for the animation, which utilizes at least a subset of the common context, is compared with output of a computational model for the animation. The computational model has a first set of logic. The data-driven model has a second set of logic that has less logic than the first set of logic. In addition, an error between the computational model and the data-driven model is computed.
Background
BACKGROUND(1) 1. Field(2) This disclosure generally relates to the field of computer graphics. More particularly, the disclosure relates to analysis of animation data.(3) 2. General Background(4) Current computer graphics approaches are utilized to modify shapes of objects, characters, etc. For example, a graphics artist may wish to direct the deformation of an object so that the object simulates the movement of cloth. However, to do so is often cumbersome or difficult since many approaches are either slow or limiting for the artist to control. Instead, the artist may employ a data-driven approach, which derives its results from a set of example data. Data-driven approaches are faster and more artistically-driven since the artist is providing the example data. However, such approaches often involve utilizing a set of data that is manually provided by a user as an example set of data. That set of data may or may not be ideal in obtaining the desired result. In other words, current approaches lack adequate means for selecting the data that is more useful rather than the data that is less useful. Further, such manual approaches are often subjective and difficult to quantify.SUMMARY(5) In one aspect of the disclosure, a process is provided. The process statistically samples a set of animation data for an element in an animation to obtain a common context. The common context is a subset of a plurality of frames of the set of animation data. Further, the process compares output of
Claims
1. A method comprising: statistically sampling, at a server, a set of animation data for an element in an animation to obtain a common context, the common context being a subset of a plurality of frames of the set of animation data; segmenting, at the server, the plurality of frames of the set of animation data into a first subset and a second subset; sending, from the server to a first client, a first data set corresponding to the first subset; sending, from the server to a second client, a second data set corresponding to the second subset; receiving, from the first client at the server, a first error that results from a first comparison in the first data set of the first subset of output of a data-driven model for the animation, which utilizes at least a subset of the common context, with output of a computational model for the animation, the computational model having a first set of logic, the data-driven model having a second set of logic that has less logic than the first set of logic; and receiving, from the second client at the server, a second error that results from a second comparison in the second data set of the second subset of output of the data-driven model for the animation, which utilizes at least the subset of the common context, with output of the computational model for the animation.
6. A system comprising: a statistical sampling module that statistically samples a set of animation data for an element in an animation to obtain a common context, the common context being a subset of a plurality of frames of the set of animation data; a segmentation module that segments the plurality of frames of the set of animation data into a first subset and a second subset; a transmission module that sends the first subset to a first client and the second subset to a second client; and a reception module that (i) receives, from the first client, a first error that results from a first comparison in the first subset of output of a data-driven model for the animation, which utilizes at least a subset of the common context, with output of a computational model for the animation and (ii) receives, from the second client at the server, a second error that results from a second comparison in the second subset of output of the data-driven model for the animation, which utilizes at least the subset of the common context, with output of the computational model for the animation, the computational model having a first set of logic, the data-driven model having a second set of logic that has less logic than the first set of logic.