CaltechAUTHORS
  A Caltech Library Service

A Rotation Invariant Latent Factor Model for Moveme Discovery from Static Poses

Ruggero Ronchi, Matteo and Kim, Joon Sik and Yue, Yisong (2016) A Rotation Invariant Latent Factor Model for Moveme Discovery from Static Poses. In: 16th IEEE International Conference on Data Mining (ICDM). IEEE International Conference on Data Mining. IEEE , Piscataway, NJ, pp. 1179-1184. ISBN 978-1-5090-5473-2. http://resolver.caltech.edu/CaltechAUTHORS:20170616-103109457

[img] PDF - Submitted Version
Creative Commons Attribution Share Alike.

3508Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20170616-103109457

Abstract

We tackle the problem of learning a rotation invariant latent factor model when the training data is comprised of lower-dimensional projections of the original feature space. The main goal is the discovery of a set of 3-D bases poses that can characterize the manifold of primitive human motions, or movemes, from a training set of 2-D projected poses obtained from still images taken at various camera angles. The proposed technique for basis discovery is data-driven rather than hand-designed. The learned representation is rotation invariant, and can reconstruct any training instance from multiple viewing angles. We apply our method to modeling human poses in sports (via the Leeds Sports Dataset), and demonstrate the effectiveness of the learned bases in a range of applications such as activity classification, inference of dynamics from a single frame, and synthetic representation of movements.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.1109/ICDM.2016.0156DOIArticle
http://ieeexplore.ieee.org/document/7837969/PublisherArticle
https://arxiv.org/abs/1609.07495arXivDiscussion Paper
Additional Information:© 2016 IEEE. Date Added to IEEE Xplore: 02 February 2017.
Record Number:CaltechAUTHORS:20170616-103109457
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20170616-103109457
Official Citation:M. R. Ronchi, J. S. Kim and Y. Yue, "A Rotation Invariant Latent Factor Model for Moveme Discovery from Static Poses," 2016 IEEE 16th International Conference on Data Mining (ICDM), Barcelona, 2016, pp. 1179-1184. doi: 10.1109/ICDM.2016.0156
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:78272
Collection:CaltechAUTHORS
Deposited By: Ruth Sustaita
Deposited On:16 Jun 2017 19:14
Last Modified:27 Mar 2019 16:01

Repository Staff Only: item control page