NeuroAnimator: fast neural network emulation and control of physics-based models

By Radek Grzeszczuk, Demetri Terzopoulos, Geoffrey E. Hinton. 1998

Subfields: ML Systems & Hardware, AI for Physical Sciences

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In ML Systems & Hardware

In AI for Physical Sciences

Citations in the corpus

Toward Causal Representation Learning (2021, background): “One interesting aspect is physics simulation with neural networks [94], which can substantially increase the efﬁciency of hand-engineered simulators [104], [143], [211], [265], [269].”
E(n) Equivariant Graph Neural Networks (2021, background): “, 2018), and physical systems simulations (Grzeszczuk et al., 1998; Watters et al., 2017).”
Towards Causal Representation Learning (2021, background): “One interesting aspect is physics simulation with neural networks [93], which can substantially increase the efficiency”
Learning to Simulate Complex Physics with Graph Networks (2020, background): “Learning simulations from data (Grzeszczuk et al., 1998) has been an important area of study with applications in physics and graphics.”
Flexible Neural Representation for Physics Prediction (2018, background): “Particles or point masses have also been used to represent more complex physical objects, with the neural network-based NeuroAnimator being one of the earliest examples to use a hierarchical particle representation for objects to advance the movement of physical objects [18].”
Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactions (2018, background): “…state-to-state predictions, often describing the state of a system as a set of compact object-descriptions that are either used as an input to the system (Battaglia et al., 2016; Chang et al., 2016; Fragkiadaki et al., 2015; Grzeszczuk et al., 1998) or for training purposes (Watters et al., 2017).”
Visual Interaction Networks: Learning a Physics Simulator from Video (2017, methodology): “The ﬁrst algorithm using this approach was the “NeuroAnimator” [12], which was able to simulate articulated bodies.”
Differentiable Physics-informed Graph Networks (2019, background): “Learning physical dynamics A class of models (Grzeszczuk et al., 1998; Battaglia et al., 2016; Chang et al., 2017; Watters et al., 2017; Sanchez-Gonzalez et al., 2018; Kipf et al., 2018) have been proposed based on the assumption that neural networks can learn complex physical interactions and…”
Graph networks as learnable physics engines for inference and control (2018, neutral): “One of our key contributions is an approach for learning physical dynamics models (Grzeszczuk et al., 1998; Fragkiadaki et al., 2015; Battaglia et al., 2016; Chang et al., 2016; Wat- 3We use the term “inference” in the sense of “abductive inference”—roughly, constructing explanations for (possibly…”
Hamiltonian Neural Networks (2019, background): “ing a neural network. The simplest way of doing this is by predicting the next state of a system given the current one. A variety of previous works have taken this path and produced excellent results [41, 14, 43, 34, 17, 6]. There are, however, a few problems with this approach. The ﬁrst problem is ”
Generating Liquid Simulations with Deformation-aware Neural Networks (2017, background): “Since initial experiments with physically-based animation and neural networks [17], a variety of new deep learning based works have been proposed to learn physical models from videos [3, 8, 41].”
Interaction Networks for Learning about Objects, Relations and Physics (2016, background): “The “NeuroAnimator” [8] was perhaps the ﬁrst quantitative approach to learning physical dynamics, by training neural networks to predict and control the state of articulated bodies.”
Interpretable Intuitive Physics Model (2018, background): “The underlying goal of most of these systems is to use physics to predict what is going to happen next [8,6,13,14,7,26,24].”
Composable controllers for physics-based character animation (2001, methodology): “Our pre-condition learning algorithm adds to the growing body of learning algorithms that have been successfully applied in the context of computer animation in recent years [14, 15].”
Optimal gait and form for animal locomotion (2009, methodology): “Simulated annealing was also used to determine the sinusoidal control pattern of swimming creatures [Grzeszczuk et al. 1998].”
Interactive motion generation from examples (2002, methodology): “These controllers can be designed speciﬁcally to accomplish particular tasks [Brogan et al. 1998; Hod-gins et al. 1995] or they can be learned automatically using statistical tools [Grzeszczuk and Terzopoulos 1995; Grzeszczuk et al. 1998; Mataric 2000].”