Human motion prediction has always been an interesting research topic in computer vision and robotics. It means forecasting human movements in the future conditioning on historical 3-dimensional human skeleton sequences. Existing predicting algorithms usually rely on extensive annotated or non-annotated motion capture data and are non-adaptive. This paper addresses the problem of few-frame human motion prediction, in the spirit of the recent progress on manifold learning. More precisely, our approach is based on the insight that achieving an accurate prediction relies on a sufficiently linear expression in the latent space from a few training data in observation space. To accomplish this, we propose Regressive Gaussian Process Latent Variable Model (RGPLVM) that introduces a novel regressive kernel function for the model training. By doing so, our model produces a linear mapping from the training data space to the latent space, while effectively transforming the prediction of human motion in physical space to the linear regression analysis in the latent space equivalent. The comparison with two learning motion prediction approaches (the state-of-the-art meta learning and the classical LSTM-3LR) demonstrate that our GPLVM significantly improves the prediction performance on various of actions in the small-sample size regime.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Xin JIN, Jia GUO, "Regressive Gaussian Process Latent Variable Model for Few-Frame Human Motion Prediction" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 10, pp. 1621-1626, October 2023, doi: 10.1587/transinf.2023PCP0001.
Abstract: Human motion prediction has always been an interesting research topic in computer vision and robotics. It means forecasting human movements in the future conditioning on historical 3-dimensional human skeleton sequences. Existing predicting algorithms usually rely on extensive annotated or non-annotated motion capture data and are non-adaptive. This paper addresses the problem of few-frame human motion prediction, in the spirit of the recent progress on manifold learning. More precisely, our approach is based on the insight that achieving an accurate prediction relies on a sufficiently linear expression in the latent space from a few training data in observation space. To accomplish this, we propose Regressive Gaussian Process Latent Variable Model (RGPLVM) that introduces a novel regressive kernel function for the model training. By doing so, our model produces a linear mapping from the training data space to the latent space, while effectively transforming the prediction of human motion in physical space to the linear regression analysis in the latent space equivalent. The comparison with two learning motion prediction approaches (the state-of-the-art meta learning and the classical LSTM-3LR) demonstrate that our GPLVM significantly improves the prediction performance on various of actions in the small-sample size regime.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2023PCP0001/_p
Copy
@ARTICLE{e106-d_10_1621,
author={Xin JIN, Jia GUO, },
journal={IEICE TRANSACTIONS on Information},
title={Regressive Gaussian Process Latent Variable Model for Few-Frame Human Motion Prediction},
year={2023},
volume={E106-D},
number={10},
pages={1621-1626},
abstract={Human motion prediction has always been an interesting research topic in computer vision and robotics. It means forecasting human movements in the future conditioning on historical 3-dimensional human skeleton sequences. Existing predicting algorithms usually rely on extensive annotated or non-annotated motion capture data and are non-adaptive. This paper addresses the problem of few-frame human motion prediction, in the spirit of the recent progress on manifold learning. More precisely, our approach is based on the insight that achieving an accurate prediction relies on a sufficiently linear expression in the latent space from a few training data in observation space. To accomplish this, we propose Regressive Gaussian Process Latent Variable Model (RGPLVM) that introduces a novel regressive kernel function for the model training. By doing so, our model produces a linear mapping from the training data space to the latent space, while effectively transforming the prediction of human motion in physical space to the linear regression analysis in the latent space equivalent. The comparison with two learning motion prediction approaches (the state-of-the-art meta learning and the classical LSTM-3LR) demonstrate that our GPLVM significantly improves the prediction performance on various of actions in the small-sample size regime.},
keywords={},
doi={10.1587/transinf.2023PCP0001},
ISSN={1745-1361},
month={October},}
Copy
TY - JOUR
TI - Regressive Gaussian Process Latent Variable Model for Few-Frame Human Motion Prediction
T2 - IEICE TRANSACTIONS on Information
SP - 1621
EP - 1626
AU - Xin JIN
AU - Jia GUO
PY - 2023
DO - 10.1587/transinf.2023PCP0001
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2023
AB - Human motion prediction has always been an interesting research topic in computer vision and robotics. It means forecasting human movements in the future conditioning on historical 3-dimensional human skeleton sequences. Existing predicting algorithms usually rely on extensive annotated or non-annotated motion capture data and are non-adaptive. This paper addresses the problem of few-frame human motion prediction, in the spirit of the recent progress on manifold learning. More precisely, our approach is based on the insight that achieving an accurate prediction relies on a sufficiently linear expression in the latent space from a few training data in observation space. To accomplish this, we propose Regressive Gaussian Process Latent Variable Model (RGPLVM) that introduces a novel regressive kernel function for the model training. By doing so, our model produces a linear mapping from the training data space to the latent space, while effectively transforming the prediction of human motion in physical space to the linear regression analysis in the latent space equivalent. The comparison with two learning motion prediction approaches (the state-of-the-art meta learning and the classical LSTM-3LR) demonstrate that our GPLVM significantly improves the prediction performance on various of actions in the small-sample size regime.
ER -