A Generative Model For Dynamic Canvas Motion
نویسندگان
چکیده
We present techniques for constructing realistic canvas and paper models and for enabling interactive dynamic canvas motion. Dynamic canvas motion means that there is a correspondence between the motion of canvas features and the motion of the models in the scene. Our artificial paper is created by simulating the physical process of creating paper with many individual fibers. To enable canvas motion, fibers are associated with each of the models in the scene. At runtime, the fibers associated with visible portions of the models and background fibers are used to construct a 2D canvas. Because fibers are “tied” to the models, the motion of canvas features corresponds to the motion of each model. This allows us to match the motion field of our dynamic 2D canvas to that of the the 3D scene exactly.
منابع مشابه
Generating Sentences Using a Dynamic Canvas
We introduce the Attentive Unsupervised Text (W)riter (AUTR), which is a word level generative model for natural language. It uses a recurrent neural network with a dynamic attention and canvas memory mechanism to iteratively construct sentences. By viewing the state of the memory at intermediate stages and where the model is placing its attention, we gain insight into how it constructs sentenc...
متن کاملA Mathematical Account of Dynamic Texture Synthesis for Probing Visual Perception
Dynamic Texture Stimulation Model: Motion Clouds We first provide an axiomatic derivation of the Motion Clouds model introduced in [1]. In particular, we show that the original definition as a spatiotemporal stationary Gaussian field could be approximated by the random aggregation of warped patterns providing a generative model. We define a dynamic texture as 1 I (x, t) = p X p✓p (x Xp Vpt)) (1...
متن کاملDynamic Canvas for Non-Photorealistic Walkthroughs
The static background paper or canvas texture usually used for non-photorealistic animation greatly impedes the sensation of motion and results in a disturbing “shower door” effect. We present a method to animate the background canvas for non-photorealistic rendering animations and walkthroughs, which greatly improves the sensation of motion and 3D “immersion”. The complex motion field induced ...
متن کاملTowards a Generative Model of Natural Motion
Towards a Generative Model of Natural Motion Cheng-Yun Karen Liu Chair of the Supervisory Committee: Professor Zoran Popović Computer Science & Engineering Understanding how animals move has fascinated scientists in many different disciplines since the ancient Greece. Today we have an unprecedented ability to study the control mechanisms of animal’s functional activities, analyze the kinematic ...
متن کاملBiologically Inspired Dynamic Textures for Probing Motion Perception
Perception is often described as a predictive process based on an optimal inference with respect to a generative model. We study here the principled construction of a generative model specifically crafted to probe motion perception. In that context, we first provide an axiomatic, biologically-driven derivation of the model. This model synthesizes random dynamic textures which are defined by sta...
متن کامل