KeypointPoseAttribute - Extended-Object-Detection-ROS/wiki_english GitHub Wiki

KeypointPoseAttribute

This attribute conducts PnP-pose estimation for extracted keypoints.

Modes Accuracy assessment 3D-translation 3D-rotation Contour extraction Additional info
E - ✔ ✔ ❌ None

1. Modes

1.1. Detect

Not implemented

1.2. Check

Not implemented

1.3. Extract

Extracts 3d pose for keypoints.

2. XML-description

2.1. Common parameters

  • Name (string, must be set) attribute unique name
  • Type (string, must be "KeypointPose") attribute type
  • Weight (double, default: 1) attribute weight
  • Probability (double, default: 0.75) acceptable detection accuracy, used in Detect mode.
  • Contour (bool, default: true) Returns the contour of the attribute if true.

2.2. Special parameters

  • keypoints (list of strings, must be set) Names of keypoints to use in calculation
  • poses (list of doubles, must be set) 3D poses for keypoints, size must be as 3 times size keypoints

2.3. Example

<?xml version="1.0" ?>

<AttributeLib>
    
    <Attribute Name="OAK_OpenPose" Type="rossubcriberopenposeraw" topic="/robot/sensors/oakd/net/raw_output" timelag="1" Probability="0.01"  kpt_score_th="0.15" paf_sample_th="0.4" min_paf_score_th="0.2" num_paf_samples="15" />

    <Attribute Name="HeadDirection" Type="KeypointPose" keypoints="nose right_eye left_eye right_ear left_ear" poses="0 0 0 -0.05 -0.025 0.035 -0.05 0.025 0.035 -0.35 -0.11 0.75 -0.35 0.11 0.75"/> 
    
</AttributeLib>

<SimpleObjectBase>  
      
    <SimpleObject Name="OpenPose" ID="2" Probability="0.01">              
        <Attribute Type="Detect">OAK_OpenPose</Attribute>            
        <Attribute Type="Extract">HeadDirection</Attribute>
    </SimpleObject>        
    
</SimpleObjectBase>
⚠️ **GitHub.com Fallback** ⚠️