Differences

This shows you the differences between two versions of the page.

--- visual3d:documentation:pipeline:expressions:signal_functions [2024/10/30 14:11] – [Second_Derivative] wikisysop
+++ visual3d:documentation:pipeline:expressions:signal_functions [2024/11/15 19:27] (current) – [Point_Tracked_By_3Points] wikisysop
@@ Line 81: / Line 81: @@
 </code>
 ==== First_Derivative ====
+[[visual3d:documentation:pipeline:signal_commands:second_derivative#first_derivative]]
 ==== Second_Derivative ====
 [[visual3d:documentation:pipeline:signal_commands:second_derivative#second_derivative]]
@@ Line 236: / Line 236: @@
 </code>
+==== Point_Tracked_By_3Points ====
+**Point_Tracked_By_3Points(Point1, Vertex1, Vertex2, Vertex3)**
+This function was implemented to used DERIVED signals as tracking markers for a point relative to these tracking markers.
+In other words, this is mimics a LANDMARK in which a point is tracked by three TARGETS or LANDMARKS in a model.
+The difference is that a model isn't required.
+Why would this be useful?\\
+The c3d format specifies that the lowest sampling rate is the POINT rate (i.e. the marker rate)\\
+Marker based tracking, however, often has possible sampling rates that are much higher than video rates used by markerless tracking.\\
+In a situation like baseball batting or golf many users record marker and markerless synchronously and merge the files.\\
+The markers are used to track the club or bat at a higher sampling rate than the video\\
+Then the c3d format, and Visual3D's implementation, collide with this objective.\\
+When merging the two files, Visual3D automatically downsamples the POINT rate to the ROTATION rate.\\
+To compensate (at least somewhat), Visual3D stores a copy of the POINT data as a DERIVED signal at the original POINT rate\\
+The data is preserved, but the model and model based items cannot refer to the DERIVED signals\\
+The workaround is to perform some calculations only on the DERIVED signals independent of any model.\\
+**As as aside, for some applications, using only the marker data will result in faster processing and smaller cmz files.**\\
+Example:
+locate a point in the model relative to 3 markers\\
+then track this point in the movement trial
+<code>
+Evaluate_Expression
+/EXPRESSION=Point_Tracked_By_3Points(Point_Relative_To_3Points(MODEL::TARGET::FaceBot, MODEL::TARGET::Shaft4, MODEL::TARGET::Shaft5, MODEL::TARGET::Shaft6),
+ TARGET::ORIGINAL::Shaft4, TARGET::ORIGINAL::Shaft5, TARGET::ORIGINAL::Shaft6)
+/RESULT_TYPES=DERIVED
+/RESULT_FOLDERS=CLUBFACE
+/RESULT_NAME=FaceBot
+;
+</code>
 ==== Snip ====