...
首页> 外文期刊>ACM Transactions on Graphics >Relationship Templates for Creating Scene Variations
【24h】

Relationship Templates for Creating Scene Variations

机译:用于创建场景变体的关系模板

获取原文
获取原文并翻译 | 示例
   

获取外文期刊封面封底 >>

       

摘要

We propose a novel example-based approach to synthesize scenesrnwith complex relations, e.g., when one object is ‘hooked’, ‘surrounded’,rn‘contained’ or ‘tucked into’ another object. Existing relationshiprndescriptors used in automatic scene synthesis methods arernbased on contacts or relative vectors connecting the object centers.rnSuch descriptors do not fully capture the geometry of spatial interactions,rnand therefore cannot describe complex relationships. Ourrnidea is to enrich the description of spatial relations between objectrnsurfaces by encoding the geometry of the open space around objects,rnand use this as a template for fitting novel objects. To this end, wernintroduce relationship templates as descriptors of complex relationships;rnthey are computed from an example scene and combine therninteraction bisector surface (IBS) with a novel feature called thernspace coverage feature (SCF), which encodes the open space in thernfrequency domain. New variations of a scene can be synthesized efficientlyrnby fitting novel objects to the template. Our method greatlyrnenhances existing automatic scene synthesis approaches by allowingrnthem to handle complex relationships, as validated by our user studies.rnThe proposed method generalizes well, as it can form complexrnrelationships with objects that have a topology and geometry veryrndifferent from the example scene.
机译:我们提出了一种基于示例的新颖方法来合成具有复杂关系的场景,例如,当一个对象被“钩住”,“环绕”,“包含”或“塞进”另一个对象时。自动场景合成方法中使用的现有关系描述符基于连接对象中心的接触或相对矢量。此类描述符无法完全捕获空间相互作用的几何结构,因此无法描述复杂的关系。 Ourrnidea将通过编码对象周围开放空间的几何形状来丰富对象表面之间空间关系的描述,并将其用作适合新颖对象的模板。为此,我们引入了关系模板作为复杂关系的描述符;它们是从示例场景中计算出来的,并将交互平分面(IBS)与称为“空间覆盖范围特征(SCF)”的新颖特征相结合,该特征在频域中对开放空间进行了编码。通过将新颖的对象拟合到模板,可以有效地合成场景的新变化。正如我们的用户研究所证实的那样,我们的方法通过允许它们处理复杂的关系极大地增强了现有的自动场景综合方法。该方法具有很好的概括性,因为它可以与拓扑结构和几何形状与示例场景非常不同的对象形成复杂的关系。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号