Interactive motion generation from examples

By Okan Arikan, D.A. Forsyth. ACM Transactions on Graphics, 2002

743 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/77a17c7e8f7341cfde92f63a04ea484d2b85881b

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

Citations in the corpus

Generalizing motion edits with Gaussian processes (2009, background): “Motion graphs encode valid rearrangements [Kovar et al. 2002; Arikan and Forsyth 2002; Lee et al. 2002].”
Style translation for human motion (2005, methodology): “Database approaches assemble short subsequences to synthesize motions that match user specifications, such as a desired path on the ground or a list of actions to perform [Kovar et al. 2002a; Lee et al. 2002; Arikan et al. 2003; Arikan and Forsyth 2002].”
Bailando: 3D Dance Generation by Actor-Critic GPT with Choreographic Memory (2022, background): “As revealed in Figure 7, the first and the second rows are 3D poses decoded from p0 = [4, 4] and p1 = [5, 5], respectively, where the former one is doing right leg lifting and the latter is right bicep curl.”
Two Distributed-State Models For Generating High-Dimensional Time Series (2011, methodology): “Perhaps the simplest way to generate new motion sequences based on data isto sensibly concatenate short examples from a motion database to meet sparse user-specified constraints (Tanco and Hilton, 2000; Arikan and Forsyth, 2002; Kovar et al., 2002; Lee et al., 2002;Arikan et al., 2003).”
Learning physics-based motion style with nonlinear inverse optimization (2005, background): “…modeling physics and style, learning simple models of style from examples has recently been an extremely active and productive area of research [Arikan and Forsyth 2002; Arikan et al. 2003; Brand and Hertzmann 2000; Grochow et al. 2004; Kovar et al. 2002; Kovar and Gleicher 2004; Lee et al.…”
Mode-adaptive neural networks for quadruped motion control (2018, background): “Data structures such as motion graphs [Arikan and Forsyth 2002; Kovar et al. 2002; Lee et al. 2002] are introduced to synthesize continuous character movements from unstructured motion capture data.”
Synthesis and evaluation of linear motion transitions (2008, methodology): “Other researchers [Arikan and Forsyth 2002; Arikan et al. 2003; Kovar and Gleicher 2003] specify transitions by the center of the blend; that is, the speci.ed frames are the midpoint in the blending interval.”
Interactive control of avatars animated with human motion data (2002, background): “Arikan and Forsyth [2002] use a hierarchy of graphs to represent connectivity of a motion database and perform randomized search to identify motions that satisfy user constraints such as motion duration and pose of the body at given keyframes.”
Motion texture: a two-level statistical model for character motion synthesis (2002, background): “More recently, several approaches have been proposed to interactively synthesize human motion by reordering the preprocessed motion capture data [21, 1, 22].”
Style-based inverse kinematics (2004, background): “Many authors have described systems for producing new sequences of movements from examples, either by direct copying and blending of poses [Arikan and Forsyth 2002; Arikan et al. 2003; Kovar et al. 2002; Lee et al. 2002; Pullen and Bregler 2002] or by learning a likelihood function over sequences…”
Catch & Carry (2019, neutral): “For generalized locomotion, various kinematic approaches sequentially set the pose of the body according to motion capture snippets [Lee et al. 2002; Arikan and Forsyth 2002; Kovar et al. 2008], with behavioral sequencing resulting from approaches ranging from high-level planning [Levine et al.…”
On Human Motion Prediction Using Recurrent Neural Networks (2017, background): “Arikan and Forsyth [2] collapse full sequences into nodes in a directed graph, connected with possible transitions between them, and in [24] cluster trees improve the path availability.”
Performance animation from low-dimensional control signals (2005, methodology): “2002; Brand and Hertzmann 2000], reordering motion clips employing a motion graph [Arikan and Forsyth 2002; Kovar et al. 2002; Lee et al. 2002; Pullen and Bregler 2002; Arikan et al. 2003] and interpolating motion to create new sequences [Guo and Roberge 1996; Wiley and Hahn 1997; Rose et al.”
Motion synthesis from annotations (2003, methodology): “We also perform a smoothing as in [Arikan and Forsyth 2002] to get rid of offensive small discontinuities.”