Difference between revisions of "Using Tracking Data from Depth Sensors"

From iPi Docs
Jump to: navigation, search
m
 
(12 intermediate revisions by 2 users not shown)
Line 1: Line 1:
{{Note|Depth sensors provide body tracking data based on embedded tracking algorithms. The accuracy of such tracking is rather low and far below requirements of professional-level motion capture solutions. Tracking provided by [[iPi Mocap Studio]] is based on iPi Soft's algorithms and do not use tracking data from sensors. However, sensor's tracking can be effectively used to improve workflow.}}  
+
=== Overview ===
 +
{{Note|Depth sensors provide body tracking data based on native tracking algorithms. The accuracy of such tracking is not enough for professional-level motion capture solutions. Tracking provided by [[iPi Mocap Studio]] is based on iPi Soft's algorithms and do not use tracking data from sensors. However, sensor's tracking can be effectively used to improve workflow.}}  
  
 
We record and use tracking data from [[Depth_Sensors_Comparison|Kinect 2]] sensors. The data are used for:
 
We record and use tracking data from [[Depth_Sensors_Comparison|Kinect 2]] sensors. The data are used for:
Line 7: Line 8:
 
* Detecting T-Poses and creation of takes
 
* Detecting T-Poses and creation of takes
 
* Setting initial pose for detected T-poses
 
* Setting initial pose for detected T-poses
* Setting inital pose for first frame of [[Region-of-Interest]]
+
* Setting initial pose for the beginning frame of the [[Region-of-Interest]]
* Warning on presense of people in sensor view during background recording
+
* Warning on presence of people in sensor's view during recording of the background
  
 
=== Recording Sensor Tracking Data ===
 
=== Recording Sensor Tracking Data ===
Line 16: Line 17:
 
[[iPi Recorder]] records tracking data provided by depth sensor for further use in [[iPi Mocap Studio]].  
 
[[iPi Recorder]] records tracking data provided by depth sensor for further use in [[iPi Mocap Studio]].  
  
{{Note|As soon as embedded tracking algorithms work well only when actor faces the sensor and each sensor works indepndently from others, for T-Pose and takes creation we only use data from first sensor.}}
+
{{Note|As soon as native tracking algorithms work well only when actor faces the sensor and each sensor works independently from others, for T-Pose and takes creation we only use data from the first sensor.}}
  
 
To make the most of sensors' tracking data, please follow this recommendations during recording:
 
To make the most of sensors' tracking data, please follow this recommendations during recording:
Line 22: Line 23:
 
* Strike T-Pose before each take
 
* Strike T-Pose before each take
 
* You have to stay in T-Pose at least for one second
 
* You have to stay in T-Pose at least for one second
* Face first sensor in T-Pose
+
* Face the first sensor in T-Pose
* If you have several actors, each of them should face first sensor
+
* If there are several actors, each of them should face the first sensor
  
{{Tip|In case when capture area does not allow to face the sensor, you can stay at an angle to sensor. Make sure that arms and legs are not occluded from sensor.}}
+
{{Tip|In case when capture area does not allow to face the sensor, you can stay at an angle to the sensor. Make sure that arms and legs are fully visible to the sensor.}}
  
{{Note|If you stand back to sensor, embedded tracking will detect gaze direction opposite to actual, so automatically detected inital pose will be incorrect and you'll have to manually fix this in [[iPi Mocap Studio]].}}
+
{{Note|If you stand back to a sensor, native tracking will detect gaze direction opposite to the actual, so automatically detected initial pose will be incorrect and you'll have to manually fix this in [[iPi Mocap Studio]].}}
  
 
=== Using Sensor Tracking Data ===
 
=== Using Sensor Tracking Data ===
  
==== Detecting Number of Actors and Actor Heights ====
+
==== Detecting the Number of Actors and Actor Heights ====
 
+
When creating a new project from recorded [[.iPiVideo File Extension|'''.iPiVideo''']] file, you need to specify number of actors and their heghts. These numbers are detected based on tracking data. You just need to check them during project creation, as embedded tracking algorithms are not very accurate. Correct the numbers if necessary.
+
  
 +
When creating a new project from recorded [[.iPiVideo File Extension|'''.iPiVideo''']] file, you need to specify the number of actors and their heights. Initial values for these fields are set based on tracking data. You just need to check them during project creation, as native tracking algorithms are not very accurate. Please correct the numbers if necessary.
 +
<center>
 
{|
 
{|
 
|[[Image:Create-project-set-actors-number.png|thumb|center|500px|Setting number of actors]]
 
|[[Image:Create-project-set-actors-number.png|thumb|center|500px|Setting number of actors]]
 
|[[Image:Create-project-set-actor-height.png|thumb|center|500px|Setting actor height]]
 
|[[Image:Create-project-set-actor-height.png|thumb|center|500px|Setting actor height]]
 
|}
 
|}
 +
</center>
 +
 +
{{Note|If you only have a single depth sensor, you will not have a step for selection of the number of actors, as multiple actors tracking is not supported for a single depth sensor.}}
 +
 +
{{Note|Actor's gender is not detected by sensor, so please set it manually for female actors.}}
 +
 +
==== Creating Takes and Setting Actor Poses ====
 +
[[Image:Create-project-set-takes-and-poses.png|thumb|450px|Takes created automatically]]
 +
* On project creation takes are created based on detected T-Poses. Each T-pose is considered as the beginning of a take.
 +
* [[Region-of-Interest]] is set to the first take
 +
* Actor's pose is set based on the first sensor's tracking data in the beginning frame of the [[Region-of-Interest]]
 +
* You can manually adjust takes and the [[Region-of-Interest]]
 +
* When you change the beginning frame of the [[Region-of-Interest]], actor's pose will be set from sensor's tracking data in the beginning frame
 +
* You must perform '''Refit Pose''' in the beginning frame of [[Region-of-Interest]] before you start tracking
 +
 +
{{Tip|Pose can be set from sensor's tracking data only if you have not performed tracking or any manual editing of the pose in this frame. Otherwise pose will not be changed.}}
 +
 +
{{Note|When you change the beginning frame of the [[Region-of-Interest]], poses previously set from sensor's tracking data are not deleted. They will not affect tracking unless you perform '''Refit Pose''' in this frame. So there's no need to manually delete them.}}
 +
 +
{{Tip|To get better result, you need to adjust actor's proportions before you start tracking.}}
 +
 +
{{Note|If actor stands back to the first sensor, his gaze direction will be set opposite to the actual, and the entire pose will be incorrect. You can delete it and reset manually.}}
 +
 +
{{Tip|If pose is not accurate enough and '''Refit Pose''' does not fix it, you need to adjust pose manually to ensure it is close to actual depth data.}}

Latest revision as of 06:42, 20 November 2015

Overview

Note.png
Note: Depth sensors provide body tracking data based on native tracking algorithms. The accuracy of such tracking is not enough for professional-level motion capture solutions. Tracking provided by iPi Mocap Studio is based on iPi Soft's algorithms and do not use tracking data from sensors. However, sensor's tracking can be effectively used to improve workflow.

We record and use tracking data from Kinect 2 sensors. The data are used for:

  • Detecting number of actors
  • Detecting actor heights
  • Detecting T-Poses and creation of takes
  • Setting initial pose for detected T-poses
  • Setting initial pose for the beginning frame of the Region-of-Interest
  • Warning on presence of people in sensor's view during recording of the background

Recording Sensor Tracking Data

T-Pose in first sensor

iPi Recorder records tracking data provided by depth sensor for further use in iPi Mocap Studio.

Note.png
Note: As soon as native tracking algorithms work well only when actor faces the sensor and each sensor works independently from others, for T-Pose and takes creation we only use data from the first sensor.

To make the most of sensors' tracking data, please follow this recommendations during recording:

  • Strike T-Pose before each take
  • You have to stay in T-Pose at least for one second
  • Face the first sensor in T-Pose
  • If there are several actors, each of them should face the first sensor
Tip.png
Tip: In case when capture area does not allow to face the sensor, you can stay at an angle to the sensor. Make sure that arms and legs are fully visible to the sensor.
Note.png
Note: If you stand back to a sensor, native tracking will detect gaze direction opposite to the actual, so automatically detected initial pose will be incorrect and you'll have to manually fix this in iPi Mocap Studio.

Using Sensor Tracking Data

Detecting the Number of Actors and Actor Heights

When creating a new project from recorded .iPiVideo file, you need to specify the number of actors and their heights. Initial values for these fields are set based on tracking data. You just need to check them during project creation, as native tracking algorithms are not very accurate. Please correct the numbers if necessary.

Setting number of actors
Setting actor height
Note.png
Note: If you only have a single depth sensor, you will not have a step for selection of the number of actors, as multiple actors tracking is not supported for a single depth sensor.
Note.png
Note: Actor's gender is not detected by sensor, so please set it manually for female actors.

Creating Takes and Setting Actor Poses

Takes created automatically
  • On project creation takes are created based on detected T-Poses. Each T-pose is considered as the beginning of a take.
  • Region-of-Interest is set to the first take
  • Actor's pose is set based on the first sensor's tracking data in the beginning frame of the Region-of-Interest
  • You can manually adjust takes and the Region-of-Interest
  • When you change the beginning frame of the Region-of-Interest, actor's pose will be set from sensor's tracking data in the beginning frame
  • You must perform Refit Pose in the beginning frame of Region-of-Interest before you start tracking
Tip.png
Tip: Pose can be set from sensor's tracking data only if you have not performed tracking or any manual editing of the pose in this frame. Otherwise pose will not be changed.
Note.png
Note: When you change the beginning frame of the Region-of-Interest, poses previously set from sensor's tracking data are not deleted. They will not affect tracking unless you perform Refit Pose in this frame. So there's no need to manually delete them.
Tip.png
Tip: To get better result, you need to adjust actor's proportions before you start tracking.
Note.png
Note: If actor stands back to the first sensor, his gaze direction will be set opposite to the actual, and the entire pose will be incorrect. You can delete it and reset manually.
Tip.png
Tip: If pose is not accurate enough and Refit Pose does not fix it, you need to adjust pose manually to ensure it is close to actual depth data.