Tools:Kinect skeletons streamer - Revision history

Frankiezafe: /* log */

2017-12-04T16:46:33Z

‎log

Frankiezafe: /* 26/11/2017 */

2017-11-27T12:24:30Z

‎26/11/2017

Frankiezafe: /* log */

2017-11-26T20:18:49Z

‎log

Frankiezafe: /* log */

2017-11-26T19:19:38Z

‎log

Frankiezafe: /* log */

2017-11-26T19:19:29Z

‎log

Frankiezafe: /* log */

2017-11-26T13:25:32Z

‎log

Frankiezafe: /* log */

2017-11-26T13:24:06Z

‎log

Frankiezafe: /* log */

2017-11-26T13:23:54Z

‎log

Frankiezafe: /* log */

2017-11-26T13:10:45Z

‎log

Frankiezafe: /* log */

2017-11-26T13:09:26Z

‎log

← Older revision		Revision as of 16:46, 4 December 2017
Line 1:		Line 1:
	== log ==		== log ==
		+
		+	=== 04/12/2017 ===
		+
		+	First test of the installation at beursschouwburg: streamer are streaming correctly, and after a correction on the viewer to detect "NaN" in the information received from the streamer, it has been possible to calibrate the 3 kinects quiet easily. Calibration is not perfect (based on skeletons, it would be a miracle), but is a good approximation of the position.
		+
		+	<html>
		+	<iframe src="https://player.vimeo.com/video/245759040?title=0&byline=0&portrait=0" width="640" height="400" frameborder="0" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>
		+	</html>

	=== 27/11/2017 ===		=== 27/11/2017 ===

@@ Line 1: / Line 1: @@
 == log ==
 === 26/11/2017 ===
@@ Line 50: / Line 63: @@
 See SkCalib folder in [https://bitbucket.org/frankiezafe/marionnettes/src src] for the code.
 === 25/11/2017 ===

@@ Line 1: / Line 1: @@
 == log ==
-'''26/11/2017''' - calibration based on successive positions of skeleton >
+=== 26/11/2017 ===
 The calibration algorithm is using 2 kinects. One is the reference, the other the calibrated. To start the process, one and only one skeleton must be present in each kinect (the same person, you got it). 500 positions of the hip center are stored (+/- 16sec). Once all the positions are collected, the calibration can be processed.
@@ Line 46: / Line 52: @@
-'''25/11/2017''' - Working on a visualing / calibrating software, called NetworkViewer in the repo. A first attempt of "auto calibration" is not giving too bad results. The orientation are not processed yet, I will work on this tomorrow.
+=== 25/11/2017 ===
 <html>
@@ Line 56: / Line 64: @@
-'''24/11/2017''' - back to my desk! La connection de 2 kinects simultanément est concluant:
+=== 24/11/2017 ===
 [[File:Kinectstreamer2kinect.png|400px]]
@@ Line 65: / Line 75: @@
-'''13/11/2017''' - premiers essais de mapping vers divers squelettes - les hanches et le "spine" posent problèmes dans la plupart des modèles...
+=== 13/11/2017 ===
 <html>
@@ Line 71: / Line 83: @@
 </html>
-'''12/11/2017''' - le calcul des orientations relatives à envoyer à blender a été tout aussi fastidieux que le reste, mais moins long. Pour ceux qui aiment la magie noire, il a fallu environ 5 heures de prises de tête pour écrire ces 2 lignes de code.
+=== 12/11/2017 ===
   if (parent != NUI_SKELETON_POSITION_COUNT) {
@@ Line 88: / Line 102: @@
 </html>
-'''10/11/2017''' - l'alignement du squelette kinect et de l'armature exportée depuis blender est maintenant automatique, plus besoin de tuner les matrices à la main
+=== 10/11/2017 ===
 [[File:Kinectstreamer-automatic-skeleton-alignment-cropped.png|400px]]
-'''08/11/2017''' - importation des matrices des os de l'armature blender dans l'application. Les orientations ne sont plus calculées par blender, ce qui posait de gros problèmespour le spin (rotations autour de l'axe alignés avec la direction de l'os).
+=== 08/11/2017 ===
 [[File:Kinectstreamer-blender-matrices-import.png|400px]]
-'''05/11/2017''' - première réussite: les orientations relatives sont calculées correctement et envoyées à blender via OSC<ref>Open Sound Control, [http://opensoundcontrol.org/ site officiel] et [[wikipedia:Open Sound Control|wikipedia]]</ref>
+=== 05/11/2017 ===
 <html>
@@ Line 102: / Line 122: @@
 </html>
-'''23/10/2017''' - code archeology, re-enabling ofxKinectNui<ref>https://github.com/sadmb/ofxKinectNui</ref>
+=== 23/10/2017 ===
 [[File:Technology-archeology-kinect.png|400px]]

@@ Line 25: / Line 25: @@
 * calibration_matrix.inverse * directions * calibration_matrix
 * and we can repeat the operation for the X axis by using a projection of the direction on the plane YZ
 Image legend:

← Older revision		Revision as of 19:19, 26 November 2017
Line 2:		Line 2:

	'''26/11/2017''' - calibration based on successive positions of skeleton >		'''26/11/2017''' - calibration based on successive positions of skeleton >
		+
		+	The calibration algorithm is using 2 kinects. One is the reference, the other the calibrated. To start the process, one and only one skeleton must be present in each kinect (the same person, you got it). 500 positions of the hip center are stored (+/- 16sec). Once all the positions are collected, the calibration can be processed.
		+
		+	It uses 3 passes:
		+
		+	* first pass: rendering the Z rotation to apply on the positions to calibrate. As Z is pointing up, this rotation is the yaw<ref name="Flight_dynamics">https://en.wikipedia.org/wiki/Flight_dynamics</ref>.
		+	* second pass: rendering the X rotation to apply on the positions once Z correction applied. As X is pointing right, this rotation is the pitch<ref name="Flight_dynamics"/>.
		+	* last pass consists in processing the average distance between reference and the positions to calibrate.
		+
		+	This transformations are stored in a matrix ready to be sent to streamers.
		+
		+	Why not compute the roll<ref name="Flight_dynamics"/>? I tried, but the poor quality of the recording was making the calibration worse than without. So I decided to drop this rotation.
		+
		+	An interesting subtlety is the way the Z and after it the X axis are processed:
		+
		+	* getting the normalised direction between pairs of reference points
		+	* getting the normalised direction between pairs of points to calibrate
		+	* projecting the 2 directions on the XY plane, by using 2d vectors
		+	* accumulating the rotation between them
		+	* averaging -> Z axis is computed -> calibration_matrix
		+	* but, before going further, directions must be rotated by the Z rotation
		+	* calibration_matrix.inverse * directions * calibration_matrix
		+	* and we can repeat the operation for the X axis by using a projection of the direction on the plane YZ
		+
		+	Image legend:

	* white points: reference, or ground truth		* white points: reference, or ground truth

← Older revision		Revision as of 13:25, 26 November 2017
Line 16:		Line 16:

	The process research has been fast thanks to [https://bitbucket.org/polymorphteam/raw/ raw]. In the viewer application, a raw "writer" have been added, witch stores each new pair positions while the calibration data retrieval. To develop the algorithm, I started a new application that loads these positions via a raw "reader". It has been easy to develop the algo on stored data instead of having to stops the viewer, change the algo, restart the app, wait for the kinect to connect, set the roles and moves in front of the kinects to generate the data.		The process research has been fast thanks to [https://bitbucket.org/polymorphteam/raw/ raw]. In the viewer application, a raw "writer" have been added, witch stores each new pair positions while the calibration data retrieval. To develop the algorithm, I started a new application that loads these positions via a raw "reader". It has been easy to develop the algo on stored data instead of having to stops the viewer, change the algo, restart the app, wait for the kinect to connect, set the roles and moves in front of the kinects to generate the data.
		+
		+	See SkCalib folder in [https://bitbucket.org/frankiezafe/marionnettes/src src] for the code.

← Older revision		Revision as of 13:24, 26 November 2017
Line 16:		Line 16:

	The process research has been fast thanks to [https://bitbucket.org/polymorphteam/raw/ raw]. In the viewer application, a raw "writer" have been added, witch stores each new pair positions while the calibration data retrieval. To develop the algorithm, I started a new application that loads these positions via a raw "reader". It has been easy to develop the algo on stored data instead of having to stops the viewer, change the algo, restart the app, wait for the kinect to connect, set the roles and moves in front of the kinects to generate the data.		The process research has been fast thanks to [https://bitbucket.org/polymorphteam/raw/ raw]. In the viewer application, a raw "writer" have been added, witch stores each new pair positions while the calibration data retrieval. To develop the algorithm, I started a new application that loads these positions via a raw "reader". It has been easy to develop the algo on stored data instead of having to stops the viewer, change the algo, restart the app, wait for the kinect to connect, set the roles and moves in front of the kinects to generate the data.
		+

	'''25/11/2017''' - Working on a visualing / calibrating software, called NetworkViewer in the repo. A first attempt of "auto calibration" is not giving too bad results. The orientation are not processed yet, I will work on this tomorrow.		'''25/11/2017''' - Working on a visualing / calibrating software, called NetworkViewer in the repo. A first attempt of "auto calibration" is not giving too bad results. The orientation are not processed yet, I will work on this tomorrow.

@@ Line 14: / Line 14: @@
 $wgGallerier->addGallery( array( "path" => "20171126_kinect_calibration/", "width" => 200 ) );
 }</script>
 '''25/11/2017''' - Working on a visualing / calibrating software, called NetworkViewer in the repo. A first attempt of "auto calibration" is not giving too bad results. The orientation are not processed yet, I will work on this tomorrow.

@@ Line 7: / Line 7: @@
 * cyan points not linked together: calibrated points rotated, before translation
 * cyan points linked: calibrated points after full matrix correction
 <script> {