2.5D cartoon models are methods to simulate three-dimensional (3D)-like movements, such as out-of-plane rotation, from two-dimensional (2D) shapes in different views. However, cartoon objects and characters have several distorted parts which do not correspond to any real 3D positions (e.g., Mickey Mouse's ears), that implies that existing systems are not suitable for designing such representations. Hence, we formulate it as a view-dependent deformation (VDD) problem, which has been proposed in the field of 3D character animation. The distortions in an arbitrary viewpoint are automatically obtained by blending the user-specified 2D shapes of key views. This model is simple enough to easily implement in an existing animation system. Several examples demonstrate the robustness of our method over previous methods. In addition, we conduct a user study and confirm that the proposed system is effective for animating classic cartoon characters.
翻译:2.5D卡通模型是模拟三维(3D)类似运动的方法,例如从不同观点的二维(2D)形状中模拟三维(3D)运动,例如飞机外旋转。但是,卡通对象和人物有几个扭曲的部分,这些部分与任何真正的三维位置(如米奇·老鼠耳朵)不相符,这意味着现有系统不适合设计这种表达方式。因此,我们把它作为3D字符动画领域提议的视向变形(VDDD)问题来拟订。任意观点的扭曲通过混合用户指定的2D关键视图的形状而自动获得。这一模型非常简单,很容易在现有的动画系统中实施。若干例子表明我们的方法比以往的方法更健全。此外,我们进行用户研究,并确认拟议的系统对模拟经典漫画字符有效。