Automated extraction and parameterization of motions in large data sets

By Lucas Kovar, Michael Gleicher. ACM Transactions on Graphics, 2004

565 citations recorded by the source corpus.

Large motion data sets often contain many variants of the same kind of motion, but without appropriate tools it is difficult to fully exploit this fact. This paper provides automated methods for identifying logically similar motions in a data set and using them to build a continuous and intuitively parameterized space of motions. To find logically similar motions that are numerically dissimilar, our search method employs a novel distance metric to find "close" motions and then uses them as intermediaries to find more distant motions. Search queries are answered at interactive speeds through a precomputation that compactly represents all possibly similar motion segments. Once a set of related motions has been extracted, we automatically register them and apply blending techniques to create a continuous space of motions. Given a function that defines relevant motion parameters, we present a method for extracting motions from this space that accurately possess new parameters requested by the user. Our algorithm extends previous work by explicitly constraining blend weights to reasonable values and having a run-time cost that is nearly independent of the number of example motions. We present experimental results on a test data set of 37,000 frames, or about ten minutes of motion sampled at 60 Hz.

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

View Paper: https://www.semanticscholar.org/paper/6fd3b901efc8acf955c89ba8cfd3c8b1cd8f4252

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, methodology): “Perhaps the simplest example is motion blending [Rose et al. 1998; Kovar and Gleicher 2004].”
Coupled Action Recognition and Pose Estimation from Multiple Views (2012, background): “Previous works have also shown that semantically similar motions belonging to the same action are not necessarily numerically similar (Kovar and Gleicher 2004; Müller et al. 2005); by encoding the pose in a relative manner, it is easier to capture semantic similarity (Müller et al.”
Documentation Mocap Database HDM05 (2007, methodology): “Applying editing, morphing, and blending techniques for the creation of new, realistic motions from prerecorded motion clips has become an active ﬁeld of research [2, 8, 11, 12, 16, 19].”
Two-person interaction detection using body-pose features and multiple instance learning (2012, background): “It is interesting to evaluate body-pose features motivated from motion capture data [20, 12, 21] using tracked skeletons from a single depth sensor.”
Performance animation from low-dimensional control signals (2005, methodology): “…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. 1998; Kovar and Gleicher 2004].”
Pose-based human action recognition via sparse representation in dissimilarity space (2014, methodology): “A naïve approach would be to use directly the coordinates of corresponding frames but as shown in [18,21], even with higher-o rder features such as skeleton models, semantic and numerica l similarity are not necessarily in line.”
Real-Time Gesture Recognition from Depth Data through Key Poses Learning and Decision Forests (2012, unknown): “angles, [18], [19] and body joints [20], [21].”
Semi-supervised time series classification (2006, background): “For example in the context of motion capture indexing, Kovar and Gleicher [ 25 ] noted that they “…add robustness to the search by concentrating on finding these closer motions and then using them as new queries in order to find more distant motions”.”
Does Human Action Recognition Benefit from Pose Estimation? (2011, background): “2 Pose-based features One of the biggest challenges of using posed-based features is that semantically similar motions may not necessarily be numerically similar [15, 22].”
Style-based inverse kinematics (2004, methodology): “Our work builds on previous example-based IK systems [ElK-oura and Singh 2003; Kovar and Gleicher 2004; Rose III et al. 2001; Wiley and Hahn 1997].”
Planning algorithms (2006, background): “Such primitives could be extracted, for example, from motion capture data [35, 556].”