Anipose Configuration File Parameters

There are many parameters in the Anipose config.toml file that can be changed according to your objectives and preferences when analyzing your data, so it is important to familiarize youself with them. These parameters pertain to project setup, camera calibration, filtering, triangulation, and angle calculation. Descriptions of each of these parameters are provided below.

Parameters for Setting up the Project

project: Name of the Anipose project (do not edit).
model_folder: This should specify the path to the DeepLabCut project folder. If the DeepLabCut folder is moved, the path should be changed accordingly.
nesting: Specifies the number of folders that are nested in structure.
video_extension: Specifies the file extension of the calibration videos.

Parameters for Calibration

board_type: Specifies the type of board used for calibration ("checkerboard", "charuco").
board_size: Width and height of grid. Should be the number of inner corners per side for checkerboard, or squares for charuco.
board_marker_bits: If ArUco or ChArUco boards are used for calibration, specifies the number of bits in the markers.
board_marker_dict_number: If ArUco or ChArUco boards are used for calibration, specifies the number of markers in the dictionary.
board_marker_length: Specifies the length of marker side.*
board_marker_separation_length: If ArUco boards are used for calibration, specifies the length of marker separation.*
board_square_side_length: If chArUco or checkerboards are used for calibration, specifies the, square side length.*
animal_calibration: Set to true if an animal was used for camera calibration.
fisheye: Set to true if the videos were taken using fisheye lens. Default set to false.

Parameters for 2D Filtering

enabled: If true, enables 2D filtering of the data.
type: Specifies the type of filter to use on the data ("viterbi" or "autoencoder").
score_threshold: Score below which labels are determined erroneous.
n_back:
medfilt: Specifies the length of the median filter.
offset_threshold: Specifies the offset from median filter to count as a jump.
spline: If true, interpolates using cubic spline instead of linear interpolation.
autoencoder_path: If the filter type is "autoencoder", specifies the path to the autoencoder file relative to config.toml.
multiprocess:

Parameters for Labeling

scheme: Defines the labeling scheme for the keypoints.

Parameters for Triangulation

triangulate: = Enables triangulation if true. Default is set to true.
cam_regex: = Regular expression to specify the camera names (ex: cam_regex = '_([A-Z])$').
optim: If true, enables optimization and applies 3D filters.
constraints: Pairs of joints to impose smoothness and spatial constraints between.
scale_smooth: Strength of enforcement of the smoothing constraints.
scale_length: Strength of enforcement of the spatial constraints.
score_threshold: Score below which labels are determined erroneous for the 3D filters.

Parameters for Angle Calculation

angle: Specifies a list of three keypoints to compute the angle between. There can any number of angle parameters, and the variable name of each one should correspond to the name of the angle (decided by you). When the angles are written to a csv file, the column headers correspond to the names you specified for each list of three keypoints. There is also the option to compute the angle for one of three types of rotations associated with the three keypoints. This can be done by specifying the string 'flex', 'axis', or 'cross-axis' as the first element in the list of angles. The following three elements in the list are still the three keypoints. If no rotation type is specified, the default rotation type is 'flex', the flexion-exension angle associated with the three keypoints.