objects
Last updated
Last updated
In the VisionAI format, an object pertains to any physical entity present within a given scene, such as a person, car, or lane marking in the real world. An object key contains various information, including the object's name, type, and contents used to describe a physical object. It uses UUID as a key. The ${OBJECT_UUID} information denotes the state of an object within a sequence and refers to its status across multiple frames.
Hint
Objects information comprises both static and dynamic elements.
The ${OBJECT_UUID} information denotes the state of an object within a sequence and refers to its status across multiple frames. Such information comprises both static and dynamic elements.
${OBJECT_UUID}
The id of the object. It uses UUID32 as a key.
object
true, unique
name
The unique name of this object. (ex. Amy) It can be any value.
string
false
type
The object class name. (ex. people) The values used in this format must conform to the ontology of the project.
string
true
frame_intervals
This key indicates which frames of this object exist. Please refer to the example & table below.
object
true
object_data
It contains static information to describe the object, such as the annotation shapes, attributes, or matrics in a sequence. (ex. the car_color: blue.) It is the static information of the object in a sequence that will not be changed via stream(sensors) or frames. If there is no static information about this object, it is not a required item.
object
false
object_data_pointers
This object points out all attributes without the value of this object and contains all static and dynamic information separately. For example, if one car with the static color blue, and a dynamic location, it will be described in different keys in a object_data_pointers.
object
true
It is an array of the object which indicates all number of frames in this sequence.
frame_start
Initial frame number of the interval.
int
true
frame_end
Ending frame number of the interval.
int
true
It contains static information to describe the object, such as the annotation shapes, attributes, or matrics in a sequence. The "object_data" here reduces redundancy by storing static information consistent throughout the sequence. This item primarily focuses on the "value" within the frames.
${CONTENT_TYPE}
The information type, which is static information. (ex. vec)
object
true
name
The name of this attribute. (ex. car_color)
string
true
val
The value of this attribute. (ex. blue)
string
true
It is an array of the object’s contents on static and dynamic information which indicates all frames in this sequence. This item primarily focuses on the "type" that exists for rapidly retrieving information without the need to explore the entire set of frames.
${CONTENT_NAME}
The object information name. (ex. bbox_shape) It would be static information or dynamic information.
object
true
type
The value type of this attribute. (ex. bbox)
string
true
frame_intervals
Shows this attribute exists in which frames. Refer to the frame_intervals above. If the object attribute is static, this can be ignored.
object
true
attributes
The attributes of this content. If there is any attribute of the objects in this sequence, it is a required item. (ex. is_occlued: false)
object
false
To describe a bbox dataset with one camera sensor:
sensor: camera (#camera1)
ontology:
people
ischild - boolean (static info)
direction - front, left, right, back (dynamic info)
age - number (static info)
car
color - white, silver, blue, red, black (static info)
truck
bus
Example Code