Human body behavior recognition method based on Grassmann manifold analysis
The invention discloses a human body behavior recognition method based on Grassmann manifold analysis. The method comprises the steps: obtaining all training samples of a data set and mapping the training samples to a Grassmann manifold space; modeling the intra-class sample point distance and the i...
Saved in:
Main Authors | , , , , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
07.08.2020
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a human body behavior recognition method based on Grassmann manifold analysis. The method comprises the steps: obtaining all training samples of a data set and mapping the training samples to a Grassmann manifold space; modeling the intra-class sample point distance and the inter-class sample point distance; redefining training samples on the data set; establishing a combined learning model; and carrying out iterative solution on the combined learning model; modeling and designing a classifier model by using various characteristics; generating a virtual label of the unlabeled video through a label propagation method from the labeled and unlabeled behavior videos based on the custom graph model, and revealing the correlation of the feature data by using multi-manifold analysis. For each type of features, the local structure consistency of neighbor data points is independently reserved, the global consistency of multiple feature data points is used in a training set to predict label data |
---|---|
Bibliography: | Application Number: CN202010293342 |