Recognizing Human Actions Using 3D Skeletal Information and CNNs
| Autor: | Apostolos Maniatis, Antonios Papadakis, Eirini Mathe, Evaggelos Spyrou, Phivos Mylonas, Ioannis Vernikos |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Activities of daily living
Computer science business.industry 020206 networking & telecommunications Pattern recognition Spectral domain 02 engineering and technology Convolutional neural network Motion (physics) Image (mathematics) Set (abstract data type) 3d space 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence business |
| Zdroj: | Engineering Applications of Neural Networks ISBN: 9783030202569 EANN |
| Popis: | In this paper we present an approach for the recognition of human actions targeting at activities of daily living (ADLs). Skeletal information is used to create images capturing the motion of joints in the 3D space. These images are then transformed to the spectral domain using 4 well-known image transforms. A deep Convolutional Neural Network is trained on those images. Our approach is thoroughly evaluated using a well-known, publicly available challenging dataset and for a set of actions that resembles to common ADLs, covering both cross-view and cross-subject cases. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|