NTU RGB+D Action Recognition Dataset

NTU RGB+D is a large-scale dataset for RGB-D action recognition, which is used for human activity analysis based on deep methods and RGB+D.

The content of this dataset can be divided into three categories:

• Forty daily actions (e.g. eating, reading, etc.)
• Nine health-related movements (such as sneezing, stumbling, etc.)
• Eleven interactive actions (such as hugging, boxing, etc.)