# DetectPoseKeypoints #

**MRE Plugin Class**
- Featurer

**Description**:

This plugin detects keypoints for a pose estimation. Pose estimation is very useful to detect key moments where body position is the indicator. The data generated by this plugin is intended to be further classified in a second featurer plugin that is specific to your content.

**Use Cases**:
- Trick shot detection (behind the back tennis shot)
- Player excitement (hands up in the air)
- Thrilling action of a soccer goalie blocking a shot flying in the air

**Dependencies**:
- MRE Helper libraries
- Scipy
- OpenCV

**ML Model dependencies**:
- PoseKeypointExtractionModel

**Other plugin dependencies**:
- None

**Parameter inputs**:
- minimum_confidence >> Minimum Confidence for keypoints extraction


**Output attributes**:
- keypoints >> the point positions determined by the algorithm
- frameId >> used for debug

**IAM permissions (least privilege)**:
- S3