Visual Ability
Introduction
Visual ability currently mainly refers to two modules of person detection and recognition. Api provides mainly in PersonApi and RobotApi.
Personnel information detection is a local capability. When a person is standing in front of the robot (excluding poor light conditions), the robot can detect the person in front. When the person is standing far away, both the face and the human body can be detected, and when the person is standing closer Only face information can be detected. When the person id is greater than or equal to 0, it means that the person’s face information is complete, and the person’s face photo can be obtained for recognition.
Person recognition requires the use of face photos for recognition. This ability is deprecated for legal reasons. You can use Google or Microsoft person recognition instead.
Person information
int id; //face id int angle; //face angle double distance; //distance int headSpeed; //robot head speed long latency; //data latency int facewidth; //facewidth int faceheight; //faceheight double faceAngleX; //X-axis angle of face double faceAngleY; //Y-axis angle of face double angleInView; //The angle of the person relative to the robot's head int faceX; //X-axis coordinate of face int faceY; //Y-axis coordinate of face int bodyX; //X-axis coordinate of human body int bodyY: //Y-axis coordinate of human body //The following fields are deprecated for legal reasons String remoteFaceId //deprecated int age; //deprecated String gender; //deprecated int glasses;//deprecated boolean with_body;//Does a human body exist? boolean with_face;//Does a face exist?
Person change monitoring
Method name: registerPersonListener / unregisterPersonListener
stopGetAllPersonInfo
Calling method:
PersonListener listener = new PersonListener() {
@Override
public void personChanged() {
super.personChanged();//when person changed, use "getAllPersons()" to get all people info in the robot's field of view
}
};
//register
PersonApi.getInstance().registerPersonListener(listener);
//unregister
PersonApi.getInstance().unregisterPersonListener(listener);
Parameter Description:
- listener: person information change monitoring
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |
Get all personnel information
Method name: getAllPersons
Calling method:
//get all people infos in the robot's field of view List<Person> personList = PersonApi.getInstance().getAllPersons(); //get all people infos within 1m of the robot's field of view List<Person> personList = PersonApi.getInstance().getAllPersons(1);
Parameter Description:
- maxDistance: What is the field of view of the robot, in meters
Return:
- personList: personnel information list
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |
Get a list of people who have detected a human body
Method name: getAllBodyList
Calling method:
//get all body infos in the robot's field of view List<Person> personList = PersonApi.getInstance().getAllBodyList(); //get all body infos within 1m of the robot's field of view List<Person> personList = PersonApi.getInstance().getAllBodyList(1);
Parameter Description:
- maxDistance: What is the field of view of the robot, in meters
Return:
- personList: personnel information list
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |
Get a list of people whose faces are detected
Method name: getAllFaceList
Calling method:
//get all face infos in the robot's field of view List<Person> personList = PersonApi.getInstance().getAllFaceList(); //get all face infos within 1m of the robot's field of view List<Person> personList = PersonApi.getInstance().getAllFaceList(1);
Parameter Description:
- maxDistance: What is the field of view of the robot, in meters
Return:
- personList: personnel information list
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |
a list of people who have detected complete faces
Method name: getCompleteFaceList
Calling method:
List<Person> personList = PersonApi.getInstance().getCompleteFaceList();
Return:
- personList: personnel information list
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |
Get the person who is following the focus
(this method is only valid when the focus is following)
Method name: getFocusPerson
Calling method:
Person person = PersonApi.getInstance().getFocusPerson();
Return:
- person: personnel information
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |
Get face photos
Method name: getPictureById
Calling method:
RobotApi.getInstance().getPictureById(reqId, faceId, count, new CommandListener() {
@Override
public void onResult(int result, String message) {
try {
JSONObject json = new JSONObject(message);
String status = json.optString("status"); //get photos successfully
if (Definition.RESPONSE_OK.equals(status)) {
JSONArray pictures = json.optJSONArray("pictures");
if (!TextUtils.isEmpty(pictures.optString(0))) {
//Photo storage local full path
String picturePath = pictures.optString(0);
}
}
} catch (JSONException | NullPointerException e) {
e.printStackTrace();
}
}
});
Parameter Description:
- faceID: face id, which can be obtained through local face recognition (Person id)
- count: the number of photos that need to be obtained, this parameter is currently invalid, and there is only one by default
Note: The pictures saved in this interface need to be deleted manually after use
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |
Start focus follow
Method name: startFocusFollow
Calling method:
RobotApi.getInstance().startFocusFollow(reqId, faceId, lostTimeout, maxDistance, new ActionListener() {
@Override
public void onStatusUpdate(int status, String data) {
switch (status) {
case Definition.STATUS_TRACK_TARGET_SUCCEED:
//follow the target succeed
break;
case Definition.STATUS_GUEST_LOST:
//lost target
break;
case Definition.STATUS_GUEST_FARAWAY:
//target out of range
break;
case Definition.STATUS_GUEST_APPEAR:
//target in range again
break;
}
}
@Override
public void onError(int errorCode, String errorString) {
switch (errorCode) {
case Definition.ERROR_SET_TRACK_FAILED:
case Definition.ERROR_TARGET_NOT_FOUND:
//can't find target
break;
case Definition.ACTION_RESPONSE_ALREADY_RUN:
//not ready, please stop all focus follow and try again
break;
case Definition.ACTION_RESPONSE_REQUEST_RES_ERROR:
//There are already api calls that need to control the chassis (such as guidance and navigation). Please stop before continuing to call
break;
}
}
@Override
public void onResult(int status, String responseString) {
Log.d(TAG, "startTrackPerson onResult status: " + status);
switch (status) {
case Definition.ACTION_RESPONSE_STOP_SUCCESS:
//api call "stopFocusFollow" succeed
break;
}
}
});
Parameter Description:
- faceID: face id, which can be obtained through local face recognition (Person id).
- lostTimeout：How long can we not identify the target and report the target loss status.
- maxDistance：How far is the target to report the over-range status. Unit meter.
Note1: When this api is used, it will take up chassis resources. You cannot perform any chassis operations at the same time, including navigation.
Note2: Don’t call this api repeatedly to track the same person, just start startFocusFollow again after the following is lost or stopped.
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |
Stop focus follow
Method name: stopFocusFollow
Calling method:
RobotApi.getInstance().stopFocusFollow(reqId);
Applicable Platform:
| GreetBot | Mini | Lucki | DeliverBot | BigScreenBot |
|---|---|---|---|---|
| Yes | Yes | Yes | Yes | Yes |