Point Cloud Episode Annotation
Last updated
Last updated
Point cloud Episodes Project
Point cloud Episodes (PCE) Project consists of one or many sequences of frames. Each sequence is called an episode/dataset in Supervisely.
PCE also includes Sensor fusion feature that supports video camera sensor in the Labeling Tool UI.
Project Meta (meta.json)
Project Meta contains the essential information about the project - Classes and Tags. These are defined project-wide and can be used for labeling in every episode inside the current project.
Episodes/Datasets (<EPISODE_NAME_1>, <EPISODE_NAME_2>, ...)
Episodes are the second level folders inside the project, they host a sequence of frames (point clouds), related photo context (images) and annotations.
Items/Point clouds (pointcloud)
Every .pcd file in a sequence has to be stored inside a pointcloud folder of episodes.
x
The x coordinate of the point.
y
The y coordinate of the point.
z
The z coordinate of the point.
r
The red color channel component. An 8-bit value (0-255).
g
The green color channel component. An 8-bit value (0-255)
b
The blue color channel component. An 8-bit value (0-255)
All the positional coordinates (x, y, z) are in meters. Supervisely supports all PCD encoding: ASCII, binary, binary_compressed.
The PCD file format description can be found here
Items Annotation (annotation.json)
Point cloud Episode Annotation contains the information for the entire episode including labels on all point clouds (frames) in the episode and objects. The mapping between frame numbers and point cloud names is specified in the file frame_pointcloud_map.json which guarantees the order.
An episode contains a list of objects that are used to track labels between frames. The list of objects is defined for the entire episode
Figures represent individual labels on frames. Label contains information about the geometry, frame number and object that it belongs to.
Optional fields and loading These fields are optional and are not needed when loading the project. The server can automatically fill in these fields while project is loading.
id - unique identifier of the current object
classId - unique class identifier of the current object
labelerLogin - string - the name of user who created the current figure
createdAt - string - date and time of figure creation
updatedAt - string - date and time of the last figure update
Main idea of key fields and id you can see below in Key id map file section.
Fields definitions:
description - string - (optional) - this field is used to store the text to assign to the sequence.
key - string, unique key for a given sequence (used in key_id_map.json to get the sequence ID)
tags - list of strings that will be interpreted as episode tags
objects - list of objects that may be present on the episode
frames - list of frames of which the episode consists. List contains only frames with an object from the 'objects' field
index - integer - number of the current frame
figures - list of figures in the current frame.
framesCount - integer - total number of frames in the episode
geometryType - "cuboid_3d" or other 3D geometry - class shape
Fields definitions for objects field:
key - string - unique key for a given object (used in key_id_map.json)
classTitle - string - the title of a class. It's used to identify the class shape from the meta.json file
tags - list of strings that will be interpreted as object tags (can be empty)
Fields description for figures field:
key - string - unique key for a given figure (used in key_id_map.json)
objectKey - string - unique key to link figure to object (used in key_id_map.json)
geometryType - "cuboid_3d" or other 3D geometry -class shape
geometry - geometry of the object
Description for geometry field (cuboid_3d):
position 3D vector of box center coordinates:
x - forward in the direction of the object
y - left
z - up
dimensions is a 3D vector that scales a cuboid from its local center along x,y,z:
x - width
y - length
z - height
rotation is a 3D Vector that rotates a cuboid along an axis in world space:
x - pitch
y - roll
z - yaw (direction)
Rotation values bound inside [-pi ; pi] When yaw = 0 box direction will be strict +y
The basic idea behind key-id-map is that it maps the unique identifiers of entities from Supervisely to local entities keys. It is needed for such local data manipulations as cloning entities and reassigning relations between them. Examples of entities in key_id_map.json: datasets (videos/episodes), tags, objects, figures.
objects - dictionary, where the key is a unique string, generated inside Supervisely environment to set mapping of current object in annotation, and values are unique integer ID related to the current object
figures - dictionary, where the key is a unique string, generated inside Supervisely environment to set mapping of object on current frame in annotation, and values are unique integer ID related to the current frame
videos - dictionary, where the key is unique string, generated inside Supervisely environment to set mapping of episode (dataset) in annotation, and value is a unique integer ID related to the current sequence
tags - dictionary, where the keys are unique strings, generated inside Supervisely environment to set mapping of tag on current frame in annotation, and values are a unique integer ID related to the current tag
Key - generated by python3 function uuid.uuid4().hex. The unique string. All key values and id's should be unique inside single project and can not be shared between frames\sequences.
Value - returned by server integer identifier while uploading object / figure / sequence / tag
This file stores mapping between point cloud files and annotation frames in the correct order.
Keys - frame order number Values - point cloud name (with extension)
Fields description:
name - string - Name of image file
entityId (OPTIONAL) - integer >= 1 ID of the Point Cloud in the system, that photo attached to. Doesn't require while uploading.
deviceId - string - Device ID or name.
timestamp - (OPTIONAL) - string - Time when the frame occurred in ISO 8601 format
sensorsData - Sensors data such as Pinhole camera model parameters. See wiki: Pinhole camera model and OpenCV docs for 3D reconstruction.
intrinsicMatrix - Array of number - 3x3 flatten matrix (dropped last zeros column) of intrinsic parameters in row-major order, also called camera matrix. It's used to denote camera calibration parameters. See Intrinsic parameters.
extrinsicMatrix - Array of number - 4x3 flatten matrix (dropped last zeros column) of extrinsic parameters in row-major order, also called joint rotation-translation matrix. It's used to denote the coordinate system transformations from 3D world coordinates to 3D camera coordinates. See Extrinsic_parameters.
Demo LYFT 3D dataset annotated - demo sample from Lyft dataset with labels.
Demo LYFT 3D dataset - demo sample from Lyft dataset without labels.
Demo KITTI point cloud episodes annotated - demo sample from KITTI 3D dataset with labels.
Demo KITTI point cloud episodes - demo sample from KITTI 3D dataset without labels.
