Human Pose Estimation and Activity Classification Using Machine Learning Approach
| Autor: | Poornima G. Nambiar, R. Raghul Sharan, A. K. Nandhana Davi, J. Arunnehru |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
0209 industrial biotechnology
Computer science business.industry Feature extraction 02 engineering and technology Machine learning computer.software_genre Field (computer science) k-nearest neighbors algorithm Support vector machine 020901 industrial engineering & automation 0202 electrical engineering electronic engineering information engineering Feature (machine learning) Robot 020201 artificial intelligence & image processing Artificial intelligence business Pose computer |
| Zdroj: | Advances in Intelligent Systems and Computing ISBN: 9789811524745 |
| DOI: | 10.1007/978-981-15-2475-2_11 |
| Popis: | Human pose estimation is mainly used for the purpose of training the robots to incorporate in a way which the actions are performed in reality. The human pose estimation is the highly exploring topic in the field of computer vision research. The objective of the dynamic pose estimation is to estimate the human pose in all the available datasets. It begins with mapping the skeletal coordinates and also visualizing the coordinates. The features within the coordinates are being obtained with the help of proportionate distance features and the divergence between the coordinates. The feature extraction has taken place separately for RGB and depth dataset. The computed data is being present in the same format as that of the vector space that is being allocated for the feature extractor. The classifiers such as support vector machines (SVMs), k-nearest neighbor (KNN), and the decision tree are being used. The major use of this learning process is to help the robot to train it similar to that of the human functionalities. This functionality can be used in any learning and training procedures to build the human-robotic system. The proper understanding of how every action takes place is learned in this process and thereby training any system with similar measures is simplified with this procedure. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|