ROSSubcriberOpenPoseRaw - Extended-Object-Detection-ROS/wiki_english GitHub Wiki
This attribute is made for OAK-D cameras, which could take some computations on board. Raw OpenPose neural net output is sent via ROS topics (this repo is used).
| Modes | Accuracy assessment | 3D-translation | 3D-rotation | Contour extraction | Additional info | Keypoints-labels |
|---|---|---|---|---|---|---|
| D | D | ❌ | ❌ | ❌ | None | nose, neck, right_shoulder, right_elbow, right_wrist, left_shoulder, left_elbow, left_wrist, right_hip, right_knee, right_ankle, left_hip, left_knee, left_ankle, right_eye, left_eye, right_ear, left_ear |
Detects keypoints of persons.
Not implemented
Not implemented
- Name (string, must be set) attribute unique name
- Type (string, must be "ROSSubcriberOpenPoseRaw") 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.
- topic (string, must be set) ROS topic with neural net output.
- timelag (double, default: 0) Allowed difference in timesteps between stamps of image and raw data.
- kpt_score_th (double, default: 0.15) threshold of keypoint detection
- num_paf_samples (int, default: 10) number of items in affinity fields
- paf_sample_th (double, default: 0.2) threshold of item of affinity field
- min_paf_score_th (double, default: 0.4) threshold of affinity field pair score
<?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" />
</AttributeLib>
<SimpleObjectBase>
<SimpleObject Name="OpenPose" ID="2" Probability="0.01">
<Attribute Type="Detect">OAK_OpenPose</Attribute>
</SimpleObject>
</SimpleObjectBase>