Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.

Data of human pose estimation demo

Lc00
Beginner
742 Views

Hi, when getting the raw output message there has three parameters. Eg ( a, b, c)

Lc00_0-1647089315052.png

If the a and b is the coordinate of x and y and how about the c? What the data after symbol " | " represent?

Thank you.

 

0 Kudos
1 Solution
Hairul_Intel
Moderator
660 Views

Hi Lc00,

Thank you for reaching out to us.

 

I've ran Human Pose Estimation Python Demo with the following command:

python human_pose_estimation_demo.py -m intel\human-pose-estimation-0005\FP16\human-pose-estimation-0005.xml -at ae -i vid0.mp4 -r

 

Here is the result for my demo:

demo_result.png

 

When we use '-r' flag, the demo will print the raw detection results which are the values for 'poses' and 'scores' as shown in the code from line 160 of human_pose_estimation_demo.py

line 160.png

 

As I am running the demo using '-at ae' flag, we can confirm the 'poses' and 'scores' values from the human_pose_estimation_demo.py and hpe_associative_embedding.py script as shown below:

 

From line 102 of human_pose_estimation_demo.py:

line 102.png

From line 127 of hpe_associative_embedding.py :

line 127.png

From line 334 of hpe_associative_embedding.py :

line 334.png

 

Based on the code snippet shared above, the three parameters of the ( a, b, c ) are the X coordinate, Y coordinate and followed by the key point confidence score. Meanwhile, the result after "|" symbol is the mean score of the total key point confidence score.

 

For sharing purposes, the raw detection result outputs 17 ( x, y, score ) values as I am using human-pose-estimation-0005 model. These represents 17 key points of the joints (ears, eyes, nose, shoulders, elbows, wrists, hips, knees, and ankles).

 

 

Regards,

Hairul

 

View solution in original post

0 Kudos
4 Replies
Hairul_Intel
Moderator
661 Views

Hi Lc00,

Thank you for reaching out to us.

 

I've ran Human Pose Estimation Python Demo with the following command:

python human_pose_estimation_demo.py -m intel\human-pose-estimation-0005\FP16\human-pose-estimation-0005.xml -at ae -i vid0.mp4 -r

 

Here is the result for my demo:

demo_result.png

 

When we use '-r' flag, the demo will print the raw detection results which are the values for 'poses' and 'scores' as shown in the code from line 160 of human_pose_estimation_demo.py

line 160.png

 

As I am running the demo using '-at ae' flag, we can confirm the 'poses' and 'scores' values from the human_pose_estimation_demo.py and hpe_associative_embedding.py script as shown below:

 

From line 102 of human_pose_estimation_demo.py:

line 102.png

From line 127 of hpe_associative_embedding.py :

line 127.png

From line 334 of hpe_associative_embedding.py :

line 334.png

 

Based on the code snippet shared above, the three parameters of the ( a, b, c ) are the X coordinate, Y coordinate and followed by the key point confidence score. Meanwhile, the result after "|" symbol is the mean score of the total key point confidence score.

 

For sharing purposes, the raw detection result outputs 17 ( x, y, score ) values as I am using human-pose-estimation-0005 model. These represents 17 key points of the joints (ears, eyes, nose, shoulders, elbows, wrists, hips, knees, and ankles).

 

 

Regards,

Hairul

 

0 Kudos
Lc00
Beginner
646 Views

Thank you for the information. Can I know where to find the hpe_associative_embedding.py?

 

0 Kudos
Hairul_Intel
Moderator
605 Views

Hi Lc00,

The hpe_associative_embedding.py file can be found in the following directory:

 

C:\Program Files (x86)\Intel\openvino_2021\deployment_tools\open_model_zoo\demos\common\python\models

 

 

Regards,

Hairul


0 Kudos
Hairul_Intel
Moderator
543 Views

Hi Lc00,

This thread will no longer be monitored since this issue has been resolved. If you need any additional information from Intel, please submit a new question.

 

 

Regards,

Hairul


0 Kudos
Reply