Dian Gong , University of Southern California, Los Angeles
Gérard Medioni , University of Southern California, Los Angeles
Xuemei Zhao , University of Southern California, Los Angeles
We address the problem of structure learning of human motion to recognize actions from a continuous monocular motion sequence of an arbitrary person from an arbitrary viewpoint. Human motion sequences are represented by multivariate time series in the joint trajectories space. Under this structured time series framework, we first propose Kernelized Temporal Cut (KTC), an extension of previous works on change-point detection by incorporating Hilbert space embedding of distributions, to handle the nonparametric and high dimensionality issues. Experimental results demonstrate the effectiveness of our approach, which yields realtime segmentation, and produces high action segmentation accuracy. Second, a spatio-temporal manifold framework is proposed to model the latent structure of time series data. Then an efficient spatio-temporal alignment algorithm Dynamic Manifold Warping (DMW) is proposed for time series to calculate motion similarity between action sequences (segments). Furthermore, by combining the temporal segmentation algorithm and the alignment algorithm, online action recognition can be performed by associating a few labeled examples from motion capture data. The results on human motion capture data and 3D depth sensor data demonstrate the effectiveness of the proposed approach in segmenting and recognizing motion sequences, and its ability to handle noisy and partially occluded data, in the transfer learning module.
Computer vision, Machine learning
D. Gong, G. Medioni and X. Zhao, "Structured Time Series Analysis for Human Action Segmentation and Recognition," in IEEE Transactions on Pattern Analysis & Machine Intelligence.