-
Notifications
You must be signed in to change notification settings - Fork 1
Annotations
Mateus Karvat edited this page Oct 3, 2024
·
1 revision
Adver-City follows an annotation schema compatible with OPV2V. However, due to changes in sensors, there are some minor differences.
For each viewpoints' folder, each frame has a YAML file with data taken directly from the CARLA server. The data within it is organized as follows:
RSU: false # if this agent is an RSU or not
camera0: # parameters for the frontal camera
cords: # camera coordinates under CARLA map coordinates
- 213.1973419189453 # x
- -5.52460241317749 # y
- 1.483937382698059 # z
- -0.00433349609375 # roll
- -174.12957763671875 # yaw
- -0.2315092533826828 # pitch
extrinsic: # extrinsic matrix from camera to lidar
- - 1.0
- -2.1286936818054485e-17
- -7.180017982536492e-20
- -0.9500086905572118
- - -4.959221109941733e-18
- 0.9999999999999998
- -4.757742642355514e-20
- 1.0234670639874821e-06
- - -4.90559482280122e-19
- 6.599812935430275e-20
- 1.0
- 0.44999985330774583
- - 0.0
- 0.0
- 0.0
- 1.0
intrinsic: # camera intrinsic matrix
- - 805.5356459301888
- 0.0
- 960.0
- - 0.0
- 805.5356459301888
- 540.0
- - 0.0
- 0.0
- 1.0
camera1: ... # parameters for right camera
camera2: ... # parameters for left camera
camera3: ... # parameters for back camera
ego_speed: 0.1760935088021043 # speed of this agent in km/h
lidar_pose: # lidar pose under CARLA coordinates
- 214.1405487060547 # x
- -5.427590370178223 # y
- 1.9377721548080444 # z
- -0.00433349609375 # roll
- -174.12957763671875 # yaw
- -0.2315092533826828 # pitch
plan_trajectory: # list of locations to be followed by this agent, as defined by the path planning algorithm
- - 212.44387817382812 # x
- -5.307567596435547 # y
- 9.429222033557476 # z
- - 211.8439483642578
- -5.316528797149658
- 9.429222033557476
- - 211.04403686523438
- -5.328477382659912
- 9.429222033557476
- ...
predicted_ego_pos: # agent`s localization from GNSS
- 213.65081787109375 # x
- -5.478086948394775 # y
- 0.0357675738632679 # z
- -0.00433349609375 # roll
- -174.12957763671875 # yaw
- -0.2315092533826828 # pitch
true_ego_pos: # true position of this agent
- 213.65081787109375 # x
- -5.478086948394775 # y
- 0.0357675738632679 # z
- -0.00433349609375 # roll
- -174.12957763671875 # yaw
- -0.2315092533826828 # pitch
vehicles: # list of vehicles within line of sight of this agent
1554: # id of the perceived vehicle
angle: # under CARLA map coordinate system
- 0.00028235220815986395 # roll
- 0.8933372497558594 # yaw
- -0.032047245651483536 # pitch
bp_id: vehicle.lincoln.mkz_2017 # blueprint id for this vehicle, from CARLA
center: # relative position from the center of the bounding box to the frontal axis of the vehicle
- 0.004043583292514086 # x
- 7.466280038670448e-08 # y
- 0.7188605070114136 # z
class: car # class/category of this object. Adver-City has 6 classes: car, van, truck, motorcycle, bicycle and pedestrian
color: 0, 0, 255 # integer r, g, b values for this vehicle
dist: 94.9173812866211 # distance from the agent, in meters
extent: # half length, width and height of the vehicle, in meters
- 2.4508416652679443
- 1.0641621351242065
- 0.7553732395172119
location: # position of the center in the frontal axis of the vehicle under CARLA map coordinate system
- 119.29178619384766 # x
- 4.802069187164307 # y
- 0.0340358167886734 # z
relative_angle: 175.0229148864746 # relative angle to the agent
speed: 12.787282262180959 # speed of the vehicle, in km/h, in relation to the world (not in relation to the agent)
1568: ...
1608: ...
walkers: # list of pedestrians within line of sight of this agent
1300:
angle: # under CARLA map coordinate system
- 0.0 # roll
- -0.6118773818016052 # yaw
- 0.0 # pitch
bp_id: walker.pedestrian.0024 # blueprint id for this pedestrian, from CARLA
class: walker # class/category of this object. For pedestrians, it is always `walker`
dist: 58.161373138427734
extent: # half length, width and height of the pedestrian, in meters
- 0.18767888844013214
- 0.18767888844013214
- 0.9300000071525574
location: # position of the center in the frontal axis of the vehicle under CARLA map coordinate system
- 159.15383911132812 # x
- 14.811898231506348 # y
- 1.1038998365402222 # z
relative_angle: 173.51770025491714 # relative angle to the agent
speed: 5.411598565237392 # speed of the pedestrian, in km/h, in relation to the world (not in relation to the agent)
1318: ...
1322: ...