Application Settings: Live Pipeline

Main pageApplication SettingsApplication Settings: General

In Motive, the Application Settings can be accessed under the View tab or by clicking Toolbar AppSettings 20.png icon on the main toolbar. Default Application Settings can be recovered by Reset Application Settings under the Edit Tools tab from the main Toolbar.

Info2.png

Applications Setting Pages:

Overview


Live-Pipeline settings contain camera filter settings and solver settings for obtaining 3D data in Motive. Please note that these settings are optimized by default and should provide high-quality tracking for most applications. The settings that might need to be adjusted based on the application are visible by default (i.e. not advanced).

The most commonly changed settings are...

  • Coarse/Fine IK Iterations - This helps skeletons converge to a good pose quickly when skeletons start in a difficult to track pose.
  • Minimum Rays to Start/Continue - This helps reduce false markers from semi-reflective objects when there is a lot of camera overlap. It also allows you to not track when seen by only one camera (Minimum Rays to Continue = 2).
  • Boot Skeleton Label Percentage - A lower value will allow skeletons to boot more quickly when entering the volume. A higher value will prevent untracked skeletons from attempting to track using other markers in the volume.


Info2.png

Solver settings for recorded captures:

Please note that these settings are applied only to the Live 3D data. For captures that are already recorded, you can optimize them from the properties of the corresponding TAK file.

Solver


Solver settings shown under application settings in Motive.

The solver settings control how each marker trajectory gets reconstructed into the 3D space and how rigid bodies and skeletons track. The solver is designed to work for most applications without needing to modify any settings. However, in some instances changing some settings will lead to better tracking results. The settings that may need to be changed are visible by default. There are also a large number of advanced settings that we don’t recommend changing, but the tooltips are available if needed. The settings that users may need to be changed are listed below with descriptions.

General Settings

These are general tracking settings for the solver not related to creating 3D markers or booting assets. Do not change these settings in Live mode as incorrect settings can negatively affect the tracking, this is mostly useful when optimizing 3D data for recorded captures with actors in difficult positions to track.

Coarse IK Iterations

  • What it does: This property sets the number of Coarse IK iterations, which are fast but not accurate inverse kinematic solve to place the skeleton on the associated markers.
  • When to change: Do not change this property in Live mode. In recorded captures, this property may need to be changed, under the TAK properties, if the recording(s) starts with actors who are not in standing-up positions. Sometimes in these recordings, the skeletons may not solve on the first couple frames, and in these cases, increasing this setting will allow the skeleton to converge on the first frame.

Fine IK Iterations

  • What it does: This property sets the number of Fine IK iterations, which are slow but accurate inverse kinematic solve to place the final pose of the skeleton on the associated markers. Increasing this setting may result in higher CPU usage.
  • When to change: Do not change this property in Live mode. In recorded captures, this property may need to be changed, under the TAK properties, if the recording(s) starts with actors who are not in standing-up positions or the ones that are difficult to solve. Sometimes in these recordings, the skeletons may not solve on the first couple frames, and in these cases, increasing this setting will allow the skeleton to converge on the first frame.

Trajectorizer

The Trajectorizer settings control how the 2D marker data is converted into 3D points in the calibrated volume. The trajectorizer performs reconstruction of 2D data into 3D data, and these settings control how markers are created in the 3D scene over time.

3D Threshold

  • What it does: This setting controls the maximum distance between a marker trajectory and its predicted position.
  • When to change: This setting may need to be increased when tracking extra fast assets. The default setting should track most applications. Attempt to track with default settings first, and if there are any gaps in the marker trajectories, you can incrementally increase the distance until stable tracking is achieved.

Minimum Rays to Start

  • What it does: This sets the minimum number of tracked rays that need to converge on one location to create a new marker in 3D. This is also the minimum number of calibrated cameras that see the same target marker within the 3D threshold value for them to initially get trajectorized into a 3D point.
  • When to change: For large volumes with high camera counts, increasing this value may provide more accurate and robust tracking. The default value of 3 works well with most medium and small-sized volumes. For volumes that only have two cameras, the trajectorizer will use a value of 2 even when it's not explicitly set.

Minimum Rays to Continue

  • What it does: This sets the minimum number of rays that need to converge on one location in order to continue tracking a marker that already initialized near that location. A value of 1 will use asset definitions to continue tracking markers even when a 3D marker could not have been created from the camera data without the additional asset information.
  • When to change: This is set to 1 by default. It means that Motive will continue the 3D data trajectory as long as at least one ray is obtained and the asset definition matches. When single ray tracking is not desired or for volumes with a large number of cameras change this value to 2 to utilize camera overlaps in the volume.

Active Pattern Depth

  • What it does: This setting is used for tracking active markers only, and it sets the number of frames of motion capture data used to uniquely identify the ID value of an active marker.
  • When to change: When using a large number of active tags or active pucks, this setting may need to be increased. It's recommended to use the active batch programmer when configuring multiple active components, and when each batch of active devices has been programmed, the programmer will provide a minimum active pattern depth value that should be used in Motive.

Minimum Active Count

  • What it does: The total number of rays that must contribute to an active marker before it is considered active and given an ID value.
  • When to change: Change this setting to increase the confidence in the accuracy of active marker ID values (not changed very often).

Maximum Fill Frames

  • What it does: The number of frames of data that the solver will attempt to fill if a marker goes missing for some reason. This value must be at least 1 if you are using active markers.
  • When to change: If you would like more or fewer frames to be filled when there are small gaps.

Booter

The Booter settings control when the assets start tracking, or boot, on the trajectorized 3D markers in the scene. In other words, these settings determine when rigid bodies and/or skeletons track on a set of markers.

Boot 3D Threshold

  • What it does: This controls the maximum distance between a pair of asset model markers to be considered as an edge in the label graph.
  • When to change: The default settings should work for most applications. This value may need to be increased this value to track large assets with markers that are far apart.

Boot Skeleton Label Percent

  • What it does: This sets the percentage of skeleton markers that need to be trajectorized in order to track a corresponding skeleton(s). If needed, this setting can also be configured per each asset from the corresponding asset properties using the Properties pane.
  • When to change: The default settings should work for most applications. Set this value to about 75% to help keep skeletons from booting on other markers in the volume if there are similar skeleton definitions or lots of loose markers in the scene. If you would like skeletons to boot faster when entering the volume, then you can set this value lower.

Boot Residual Percent

  • What it does: This value controls how willing an asset is to boot on to markers. A higher value will make assets boot faster when entering the volume. A lower value will stop assets from booting onto other markers when they leave the volume.
  • When to change: This property can be used in conjunction with the “Boot Skeleton Label Percent” to move between booting faster or waiting until the solver is more confident of a solution.

Cameras


This section of the application settings is used for configuring the 2D filter properties for all of the cameras.

Basic Settings

Minimum Pixel Threshold

The minimum pixel size of a 2D object, a collection of pixels grouped together, for it to be included in the Point Cloud reconstruction. All pixels must first meet the brightness threshold defined in the Cameras pane in order to be grouped as a 2D object. This can be used to filter out small reflections that are flickering in the view. The default value for the minimum pixel size is 4, which means that there must be 4 or more pixels in a group for a ray to be generated.

Circularity

The circle filter omitting non-circular reflections from a 2D camera view.
This setting sets the threshold of the circularity filter. Valid range is between 0 and 1; with 1 being a perfectly round reflection and 0 being flat. Using this 2D object filter, the software can identify marker reflections using the shape, specifically the roundness, of the group of thresholded pixels. Higher circularity setting will filter out all other reflections that are not circular. It is recommended to optimize this setting so that extraneous reflections are efficiently filtered out while not filtering out the marker reflections.
When using lower resolution cameras to capture smaller markers at a long distance, the marker reflection may appear to be more pixelated and non-circular. In this case, you may need to lower the circularity filter value for the reflection to be considered as a 2D object from the camera view. Also, this setting may need to be lowered when tracking non-spherical markers in order to avoid filtering the reflections.

Noteable Advanced Settings

Advanced Settings

Synchronizer Control

Determines how late camera frames are dealt with. Timely Delivery will drop late frames, which is ideal for real-time applications where data completeness is secondary to timeliness. Complete Delivery will hold up processing of frames when a frame is late. Automatic, which is the default and recommended setting, runs in Timely Delivery mode until it gets a non-trivial percentage of late frames, at which point it will automatically switch to Complete Delivery.

Filter Type

Toggles 2D object (Size and Roundness) filtering on or off.
This filter is very useful for filtering out extraneous reflections according to their characteristics (size and roundness) rather than blocking pixels using the masking tool or the Block Visible feature. Turn off this setting only when you want to use every 2D pixels above the brightness threshold from camera views. When there are extraneous or flickering reflections in the view, turn on the filter to specify and consider reflections only from markers. There are multiple filtering parameters to distinguish the marker reflections. Given that there are assumed marker characteristics, filtering parameters can be set. The size parameters can be defined to filter out extra-small or extra-large reflections that are most likely from extraneous sources other than markers. Non-circular reflections can be ignored assuming that all reflective markers have circular shapes. Note that even when applying the size and roundness filter, you should always Block Visible when you calibrate.

Maximum Pixel Threshold

The size filter applied with a maximum size limit of 200 pixels. Any reflections bigger than 200 pixel-size is omitted from the Point Cloud reconstruction calculation.
The maximum size of a 2D object, in pixels, in order for it to be included in point cloud reconstruction. Default is 2000 pixels which basically means that all of detected large reflections smaller than 2000 pixel-size will be included as a 2D object. Use this to filter out larger markers in a variable marker capture. For instance, if you have 4 mm markers on an actor's face and 14 mm markers on their body, use this setting to filter out the larger markers if the need arises.

Intrusion Band

The intrusion band feature allows cameras to recognize reflections that are about to be merged and filter them out before it happens. This filter occurs before the circularity filter, and these reflections are rejected before the thresholded pixels merge. This is useful for improving the accuracy of the tracking, because bright pixels from close by reflections may slightly shift the centroid locations. The intrusion band value is added to the calculated radius of detected markers to establish a boundary, and any extraneous reflections intruding the boundary is considered as the intrusion and gets omitted. When an intrusion happens, both intruding reflection and detected marker reflection will be filtered out.

Grayscale Floor

The grayscale floor setting further darkens pixels with lower brightness intensity values.

Object Margin Diameter

The object margin adds an additional margin on top of the intrusion band for filtering out merged reflections. Lowering this value will better detect close-by reflections, but may decrease the accuracy of the centroid positions as a tradeoff.

Object Skew

The number of 2D pixels a 2D object is allowed to lean.

Max Aspect Tolerance

The maximum allowable aspect tolerance (width:height) to process a 2D object.

Aspect Base

The allowable aspect tolerance for very small objects.

Aspect Step Size

The rate at which the aspect tolerance relaxes as object size increases.