pepper.framework.sensor.obj module¶

class pepper.framework.sensor.obj.Object(name, confidence, bounds, image)[source]¶

Bases: object

‘Object’ object

Parameters:	name (str) – Name of Object confidence (float) – Object Name & Bounds Confidence bounds (Bounds) – Bounds in Image Space image (AbstractImage) – Image from which Object was Recognised

bounds¶

[0, pi]}

Returns:	bounds – Object Bounding Box in View Space
Return type:	Bounds
Type:	Object Bounds in View Space {x
Type:	[-pi, +pi], y

bounds3D¶

3D bounds (for visualisation) [x,y,z]*4

Returns:	bounds3D – 3D bounds (for visualisation) [x,y,z]*4
Return type:	List[Tuple[float, float, float]]

confidence¶

Object Confidence

Returns:	confidence – Object Name & Bounds Confidence
Return type:	float

depth¶

Distance from Camera to Object

Returns:	depth – Distance from Camera to Object
Return type:	float

dict()[source]¶

Object to Dictionary

Returns:	dict – Dictionary representation of Object
Return type:	Dict

direction¶

Direction of Object in View Space (equivalent to self.bounds.center)

Returns:	direction – Direction of Object in View Space
Return type:	float, float

distance_to(obj)[source]¶

Distance from this Object to obj

Parameters:	obj (Object) –
Returns:	distance
Return type:	float

classmethod from_json(data, image)[source]¶

id¶

Object ID

Returns:	id
Return type:	int

image¶

Image associated with the observation of this Object

Returns:	image
Return type:	AbstractImage

image_bounds¶

[0, 1]}

Returns:	bounds – Object Bounding Box in Image Space
Return type:	Bounds
Type:	Object Bounds in Image Space {x
Type:	[0, 1], y

json()[source]¶

Object to JSON

Returns:	json
Return type:	JSON representation of Object

name¶

Object Name

Returns:	name – Name of Person
Return type:	str

position¶

Position of Object in Cartesian Coordinates (x,y,z), Relative to Camera

Returns:	position – Position of Object in Cartesian Coordinates (x,y,z)
Return type:	Tuple[float, float, float]

time¶

Time of Observation

Returns:	time
Return type:	float

class pepper.framework.sensor.obj.ObjectDetectionClient(target)[source]¶

Bases: object

Object Detection Client - Communicates with pepper_tensorflow Server

Parameters:	target (ObjectDetectionTarget) – Which Object Detection Server to target

classify(image)[source]¶

Classify Objects in Image

Parameters:	image (AbstractImage) – Image (Containing Objects)
Returns:	objects – Classified Objects
Return type:	List[Object]

target¶

Object Detection Target

Returns:	target
Return type:	ObjectDetectionTarget

pepper.framework.sensor.obj.getrandbits(k) → x. Generates a long int with k random bits.¶