| 摘要 |
Provided are a method, an apparatus, and a computer program product for behavior recognition according to an exemplary embodiment. The method comprises a step in which the apparatus receives sampled data related to behaviors from one or more sensor connected to the apparatus wirelessly. The sampled data is generated in the one or more sensor based on a compression sampling of the behavior data related to the behavior. The compression sampling is performed based on sampling information. The behaviors are classified at least based on the sampled data. An error associated with the classification of the behaviors is determined. The sampling information is updated or maintained based on a comparison between the error and a threshold error. The updated sampling information is used to generate updated sampling data. The updated sampled data enables reclassification of the behaviors. |
