Resources
DatasetResource
Represents the metadata of a dataset.
Properties
Property | Value | Description |
---|---|---|
dataset_id | string | Id of the dataset |
client_id | string | Id of the workspace the dataset belongs to |
dataset_name | string | Name of the dataset |
pipeline_spec | array | Labelling stages for the dataset with stage_id and edit permission. |
pipeline_status_by_label_set_id | object | Number of tasks in each label set id in each stage for each task status type |
sensor_config | object | Instance of SensorConfigResource |
uploaded_files | array | List of uploaded files |
fps | integer | Fps (Frames per second) information for video type datasets |
frame_sizes | array | List of number of points in each frame for point cloud datasets |
create_timestamp_millis | float | Timestamp at which dataset was created |
dataset_format | string | Format of uploaded file for Point Cloud datasets |
dataset_type | string | Type of dataset (images, videos, 3d) |
labelling_mode | string | Labelling mode for the dataset
|
tags | array | List of TagResource |
files | array | Array of file metadata objects. Each object represents metadata about a file being uploaded as part of the dataset. |
current_stage | string | Possible values for current stage: "FileUpload", "Processing", "Labelling", "Audit" |
current_stage_status | string | Possible values for current stage status: "waiting", "ready", "in_progress", "done" |
user_scopes | array | Array of DatasetUserScopeResource |
user_group_scopes | array | Array of DatasetUserGroupScopeResource |
Example
1
{
2
"dataset_id": "dataset_id",
3
"client_id": "client_id",
4
"create_timestamp_millis": 1591397296799,
5
"dataset_format": "json",
6
"dataset_name": "dataset_name",
7
"dataset_type": "3d",
8
"files": [
9
{
10
"file_size": 18611529.0,
11
"file_type": "application/zip",
12
"resumable_upload_url": "https://www.googleapis.com/upload/storage/v1/b/url_for_file"
13
}
14
],
15
"labelling_mode": "frame_by_frame",
16
"processing_status": "files_getting_labelled",
17
"tags": [],
18
"uploaded_files": [
19
"json_data.zip"
20
],
21
"processing_status_message": "Processing Json Files",
22
"sensor_config": {
23
"primary_sensor_id": "lidar",
24
"sensors": {
25
"lidar": {
26
"content": "path_to_lidar",
27
"sensor_type": "lidar",
28
"calibration": {},
29
"square_dimensions": [],
30
"sensor_fusion": {
31
"camera_0": {
32
"view_matrix": [
33
[1, 0, 0, 0],
34
[0, 1, 0, 0],
35
[0, 0, 0, 1],
36
[0, 0, 1, 0]
37
]
38
}
39
}
40
},
41
"camera_0": {
42
"content": "path_to_camera_0_img",
43
"sensor_type": "camera",
44
"extension": "jpg",
45
"calibration": {},
46
"square_dimensions": [],
47
"sensor_fusion": {}
48
}
49
}
50
},
51
"folder_path": "clent_id/defaultproject/dataset_id",
52
"frame_sizes": [51162, 51156, 51102, 51046, 51068, 51064, 51041, 50953, 51064, 51065, 51065],
53
"pipeline_spec": [
54
{
55
"stage_id": "Labelling",
56
"can_edit": true
57
},
58
{
59
"stage_id": "QA",
60
"can_edit": false
61
}
62
],
63
"current_stage": "Labelling",
64
"current_stage_status": "ready",
65
"pipeline_status_by_label_set_id": {
66
"default": {
67
"Labelling": {
68
"done": 0,
69
"failed": 0,
70
"in_progress": 0,
71
"ready": 0,
72
"waiting": 0
73
},
74
"QA": {
75
"done": 0,
76
"failed": 0,
77
"in_progress": 0,
78
"ready": 0,
79
"waiting": 0
80
}
81
}
82
}
83
}
Represents the metadata of a Label.
Properties
Property | Type | Description |
---|---|---|
file_id | string | File Id of the label resource |
label_type | string | Label type of label resource |
label_category_id | string | Label category Id of the label resource |
label_id | string | Label Id of the label resource |
box | array | Array of four float numbers representing the x, y coordinates, x-length, y-length of the box respectively. |
polygons | array | Array of polygons. Each polygon is an array of points. |
attributes | object | Object containing key value pairs of attribute keys and values. |
Example
1
{
2
"file_id": "000019.png",
3
"label_category_id": "bus",
4
"label_id": "bus:1",
5
"label_type": "box",
6
"attributes": {
7
"attribute2": "attribute2_value1",
8
"attribute1": "attribute_value1"
9
},
10
"box": [366, 87, 174, 116],
11
"polygons": [
12
[[338, 330], [460, 253], [567, 179], [773, 91], [862, 59]]
13
]
14
}
Represents the metadata of a file.
Properties
Property | Type | Description |
---|---|---|
file_id | string | File Id of the file resource |
Example
1
{
2
"file_id": "000009.png",
3
}
Represents a notification from Deepen AI. Currently, we support notification from Depeen AI after labelling of a dataset is complete. The url to notify should be configured at the dataset level.
Properties
Property | Type | Description |
---|---|---|
dataset_id | string | Dataset Id of the notification |
file_id | string | File Id of the notification |
notification_type | string | Represents the type of notification. Currently, we support two notification types: DatasetLabellingComplete: Sent after the dataset is completely labelled FileLabellingComplete: Send after an individual file is labelled. |
Example
1
{
2
"dataset_id": "testdataset",
3
"notification_type": "DatasetLabellingComplete"
4
}
Represents the metadata of an Issue.
Properties
Property | Type | Description |
---|---|---|
issue_id | string | Issue Id of the issue resource |
status | string | Status of the issue resource |
issue_description | string | Description of the issue resource |
file_id | string | File Id of the issue resource |
author_email | string | Email of author for issue resource |
label_category | string | Label category of the issue resource |
label_id | string | Label Id of the issue resource |
issue_creation_timestamp | Object (date) | Created time for issue resource in millis |
region | Extruded polygon defining the region for the issue. |
Example
1
{
2
"issues": [
3
{
4
"author_email": "[email protected]",
5
"issue_title": "test Issue",
6
"label_id": "test",
7
"issue_creation_timestamp": {
8
"$date": 1551194067161
9
},
10
"issue_description": "Issue Description",
11
"issue_id": "6005288",
12
"status": "to do",
13
"label_category": "test",
14
"region": {
15
"extruded_polygon": {
16
"center_z": 1.0,
17
"xy_polygon": [
18
[1.0, 2.0],
19
[3.0, 4.0],
20
[5.0, 6.0]
21
],
22
"height": 1.0
23
}
24
}
25
26
}
27
]
28
}
Represents the metadata of a task.
Properties
Property | Type | Description |
---|---|---|
task_id | string | Id of the task resource |
status | string | Status of the task resource |
client_id | string | Client_id of client |
file_id | string | File Id of the task resource |
email | string | Assigned user email for task resource |
label_category_id | string | Label category of the task resource |
sensor_id | string | Sensor Id of the task resource |
stage_id | string | Stage Id of the task resource |
label_id | string | Label Id of the task resource |
label_type | string | Label type of task resource |
task_type | string | Task type of task resource. File for a single frame Label for sequence |
dataset_id | string | dataset_id of dataset |
previous_task_ids | List | List of task ids to be completed before starting a task (Note: It will not have previous_task_ids if the particular task is 1st task) |
next_task_ids | List | List of task ids to be started after completion of a task (Note: It will not have next_task_ids if the particular task is last task) |
history | List | List of changed events containing list of changed events which contains changed_by_email, timestamp_millis, from_status, to_status (Note: It will not have history if there is no relevant history for particular task) |
Example
1
{
2
"label_tasks": [
3
{
4
"task_id": "12345",
5
"client_id": "test client",
6
"dataset_id": "test dataset",
7
"label_type": "box",
8
"label_category_id": "car",
9
"label_id": "car:1",
10
"sensor_id": "lidar",
11
"status": "ready",
12
"stage_id": "labelling",
13
"email": "[email protected]",
14
"next_task_ids": [12345],
15
"previous_task_ids": [12345678],
16
}
17
]
18
}
1
{
2
"file_tasks": [
3
{
4
"task_id": "12345",
5
"client_id": "test client",
6
"dataset_id": "test dataset",
7
"file_id": "test file",
8
"status": "ready",
9
"stage_id": "labelling",
10
"email": "[email protected]",
11
"next_task_ids": [12345],
12
"previous_task_ids": [12345678],
13
}
14
]
15
}
Represents the details of an embedded viewer.
Properties
Property | Type | Description |
---|---|---|
client_id | string | clientId of the viewer |
create_timestamp_millis | long | Created timestamp millis of viewer |
dataset_id | string | datasetId of the viewer |
jwt_token | string | Access token |
viewer_id | string | viewerId of viewer |
launching_frame_index | int | Frame id of viewer |
title | string | Name of viewer |
enable_camera_view | boolean | Flag to show camera |
expiry_timestamp_millis | long | End timestamp millis of viewer |
deletion_status | string | Deletion status of viewer |
created_by | string | User created viewer |
Example
1
{
2
"client_id": "testclient",
3
"dataset_id": "testdataset",
4
"jwt_token": "xtlz1235",
5
"viewer_id": "1234567",
6
"create_timestamp_millis": 100000000000,
7
"title": "testviewer",
8
"launching_frame_index": 1,
9
"enable_camera_view": true,
10
"expiry_timestamp_millis": 100000000000,
11
"created_by": "[email protected]"
12
}
Properties
Property | Type | Description |
---|---|---|
center_z | float | Z value of center for the extruded polygon |
xy_polygon | list(list(float)) | List of xy points for the polygon. First point is the x value and second is the y value. |
height | float | Height of the polygon |
Example
1
{
2
"extruded_polygon": {
3
"center_z": 1.0,
4
"xy_polygon": [
5
[ 1.0, 2.0 ],
6
[ 3.0, 4.0 ],
7
[ 5.0, 6.0 ]
8
],
9
"height": 1.0
10
}
11
}
Represents the details of a pipeline stage.
Properties
Property | Type | Description |
---|---|---|
stage_id | string | Id of the pipeline stage |
can_edit | boolean | Flag representing the editability of labels present in this stage. |
Example
// Some codejson
1
{
2
"stage_id": "Labelling",
3
"can_edit": "true"
4
}
Properties
Property | Type | Description |
---|---|---|
dataset_id | string | Id of the dataset |
client_id | string | Id of the client |
created_email | string | Email of author for labelling profile resource |
created_timestamp_millis | long | Timestamp during creation |
features | Object | Object with keys as label types (‘box’, ‘3d_bbox’, ‘polygon’, ‘lane’ etc) and value of type FeaturesResource |
instructions | Object | |
notification_url | string | |
pipeline_spec | Object | Pipeline stages |
profile_id | string | Id of profile |
profile_type | string | Type of profile, e.g : dataset |
updated_email | string | Email of the user who modifies it |
updated_timestamp_millis | long | Timestamp during modification |
Example
1
{
2
"dataset_id" : "datset_id",
3
"client_id" : "client_id",
4
"created_email" : "[email protected]",
5
"created_timestamp_millis" : 1590037884298,
6
"features" : {
7
"box" : {
8
"attributes" : {},
9
"label_category_ids" : []
10
},
11
"lane" : {
12
"attributes" : {},
13
"label_category_ids" : []
14
},
15
"polygon" : {
16
"attributes" : {},
17
"label_category_ids" : []
18
},
19
"point" : {
20
"attributes" : {},
21
"label_category_ids" : []
22
},
23
"3d_bbox" : {
24
"attributes" : {},
25
"label_category_ids" : [
26
"car"
27
],
28
"label_category_default_dimensions" : {
29
"car" : {
30
"width" : 2.25,
31
"height" : 2.25,
32
"length" : 2.25
33
}
34
}
35
},
36
"3d_polyline" : {
37
"attributes" : {},
38
"label_category_ids" : []
39
},
40
"3d_polygon" : {
41
"attributes" : {},
42
"label_category_ids" : []
43
},
44
"3d_point" : {
45
"attributes" : {},
46
"label_category_ids" : []
47
},
48
"3d_instance_point" : {
49
"attributes" : {},
50
"label_category_ids" : []
51
}
52
},
53
"instructions" : {
54
"text" : ""
55
},
56
"notification_url" : "",
57
"pipeline_spec" : [
58
{
59
"stage_id" : "Labelling",
60
"can_edit" : true
61
},
62
{
63
"stage_id" : "QA",
64
"can_edit" : false
65
}
66
],
67
"profile_id" : "profile_id",
68
"profile_type" : "dataset",
69
"updated_email" : "[email protected]",
70
"updated_timestamp_millis" : 1590038192070
71
}
Represents the tag details for a dataset. A tag is basically any string that can be used to identify or categories the datasets. For backward compatibility with projects, we define a special tag starting with “project:”. This will enable you to group datasets by project and you can use these tags to filter the datasets. For api calls, a sample tags field will look like below.
Example
1
{
2
"tags": ["project: testDeepenProject", "testtag"]
3
}
Properties
Property | Type | Description |
---|---|---|
email | string | Email of the user |
scopes | array | Array of user scopes |
Example
Properties
Property | Type | Description |
---|---|---|
user_group_id | string | Id of the user group |
scopes | array | Array of user scopes |
Example
1
{
2
"user_group_id" : "asdfrtfdv",
3
"scopes" : ["scope1", "scope2"]
4
}
Properties
Property | Type | Description |
---|---|---|
primary_sensor_id | string | Id of the primary sensor |
sensors | object | Object containing different sensor ids as keys with value as SensorConfigSensorDetailsResource |
Example
1
{
2
"primary_sensor_id": "lidar",
3
"sensors": {
4
"lidar": {
5
"content": "path_to_lidar"