Skip to main content

Key Points Detection

Description​

Key points detection is a tool that allows the user to detect key points in images using a pre-trained model. It is useful for various applications such as determining objects' orientation, pose estimation, and gesture recognition.

Settings​

Model Folder Path​

The model folder path is the path to the folder containing the model files. The model files are the files that are used to detect key points in the images. The model folder path is must be located on the server's machine.

See Model Folder Structure for more information about how to structure the model folder.

info

Supported Architectures​

Here are the supported frameworks and architectures:

ONNX TensorFlow 2​

Label file​

The label file is file containing the labels that will be used to label the key points. This is in JSON format.

Below is an example of a label file:

{
  "names": [
    "nose",
    "left_eye",
    "right_eye",
    "left_ear",
    "right_ear",
    "left_shoulder",
    "right_shoulder",
    "left_elbow",
    "right_elbow",
    "left_wrist",
    "right_wrist",
    "left_hip",
    "right_hip",
    "left_knee",
    "right_knee",
    "left_ankle",
    "right_ankle"
  ],
  "connections": [
    [0, 1],
    [0, 2],
    [1, 3],
    [2, 4],
    [0, 5],
    [0, 6],
    [5, 7],
    [7, 9],
    [6, 8],
    [8, 10],
    [5, 6],
    [5, 11],
    [6, 12],
    [11, 12],
    [11, 13],
    [13, 15],
    [12, 14],
    [14, 16]
  ],
  "colours": [
    [3, 16, 255],
    [255, 81, 0],
    [3, 16, 255],
    [255, 81, 0],
    [3, 16, 255],
    [255, 81, 0],
    [3, 16, 255],
    [3, 16, 255],
    [255, 81, 0],
    [255, 81, 0],
    [0, 255, 196],
    [3, 16, 255],
    [255, 81, 0],
    [0, 255, 196],
    [3, 16, 255],
    [3, 16, 255],
    [255, 81, 0],
    [255, 81, 0]
  ]
}

Where names is the list of key point names, connections is the list of connections between the key points, and colours or colors is the list of colours for each connection. The names, connections, and colours or colors are corresponding to each other.

For example, the first element in the names list is nose, the first element in the connections list is [0, 1], and the first element in the colours or colors list is [3, 16, 255]. This means that the nose key point is connected to the left_eye key point, and the colour of the connection is #FF1003 (red).

Architecture​

The architecture is the type of model that will be used to segment the images. See Supported Architecture for more information.

Bounding boxes Confidence Threshold​

The bounding boxes confidence threshold is the minimum confidence score a bounding box should have to be considered valid. Bounding boxes with confidence scores below this threshold will be discarded. A higher confidence threshold will result in fewer bounding boxes, but with higher accuracy, whereas a lower threshold will result in more bounding boxes, but with potentially lower accuracy.

Key points Confidence Threshold​

The key points confidence threshold is the minimum confidence score a key point should have to be considered valid. Key points with confidence scores below this threshold will be discarded. A higher confidence threshold will result in fewer key points, but with higher accuracy, whereas a lower threshold will result in more key points, but with potentially lower accuracy.

Advanced Settings​

Whether to enabled advanced settings. The advanced settings allow the user to have more control over the model parameters.

caution

Please use the advanced settings with caution as changing the parameters could result in unexpected behaviour.

To RGB​

Advanced Settings

Convert from BGR colours (default in Zene) to RGB colours.This is useful when using models trained on images with different channel orders.

Scale​

Advanced Settings

The scale is the scale factor that is applied to the input image. The scale factor is used to normalise the input image before it is passed to the model. The scale factor is calculated by dividing the input image by the scale factor. A higher scale factor will result in a smaller input image, while a lower scale factor will result in a larger input image.

Mean​

Advanced Settings

The mean is the mean value that is subtracted from the input image. The mean value is used to normalise the input image before it is passed to the model. The mean value is calculated by subtracting the input image by the mean value. A higher mean value will result in a darker input image, while a lower mean value will result in a brighter input image.

Input Pixel (width and height)​

Advanced Settings

The input pixel is the size of the input image. The input image is the image that is passed to the model. The user can specify the width and height of the input image.

Display Results​

Overlay Results​

Whether to draw the results on top of the image frame.

Draw Labels​

Whether to draw the results on top of the image frame.