-
Notifications
You must be signed in to change notification settings - Fork 6
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
test: detach uai Frame and Scene tests from .json file
- Loading branch information
Showing
3 changed files
with
307 additions
and
175 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
250 changes: 169 additions & 81 deletions
250
tests/test_raillabel/format/understand_ai/test_uai_frame.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,99 +1,187 @@ | ||
# Copyright DB Netz AG and contributors | ||
# SPDX-License-Identifier: Apache-2.0 | ||
|
||
import os | ||
import sys | ||
from decimal import Decimal | ||
from pathlib import Path | ||
from uuid import UUID | ||
|
||
import pytest | ||
|
||
sys.path.insert(1, str(Path(__file__).parent.parent.parent.parent)) | ||
sys.path.insert(1, str(Path(__file__).parent.parent.parent.parent.parent)) | ||
|
||
import raillabel.format.understand_ai as uai_format | ||
|
||
|
||
def _prepare_frame_data(frame_data: dict, json_data: dict) -> dict: | ||
frame_data["annotations"]["2D_BOUNDING_BOX"] = [json_data["_understand_ai_t4_format/bounding_box_2d"]] | ||
frame_data["annotations"]["2D_BOUNDING_BOX"][0]["sensor"] = json_data["_understand_ai_t4_format/sensor_reference_camera"] | ||
frame_data["annotations"]["2D_POLYLINE"] = [json_data["_understand_ai_t4_format/polyline_2d"]] | ||
frame_data["annotations"]["2D_POLYLINE"][0]["sensor"] = json_data["_understand_ai_t4_format/sensor_reference_camera"] | ||
frame_data["annotations"]["2D_POLYGON"] = [json_data["_understand_ai_t4_format/polygon_2d"]] | ||
frame_data["annotations"]["2D_POLYGON"][0]["sensor"] = json_data["_understand_ai_t4_format/sensor_reference_camera"] | ||
frame_data["annotations"]["3D_BOUNDING_BOX"] = [json_data["_understand_ai_t4_format/bounding_box_3d"]] | ||
frame_data["annotations"]["3D_BOUNDING_BOX"][0]["sensor"] = json_data["_understand_ai_t4_format/sensor_reference_lidar"] | ||
frame_data["annotations"]["3D_SEGMENTATION"] = [json_data["_understand_ai_t4_format/segmentation_3d"]] | ||
frame_data["annotations"]["3D_SEGMENTATION"][0]["sensor"] = json_data["_understand_ai_t4_format/sensor_reference_lidar"] | ||
|
||
return frame_data | ||
|
||
|
||
def _prepare_ground_truth(ground_truth: dict, json_data:dict) -> dict: | ||
ground_truth["objects"]["48c988bd-76f1-423f-b46d-7e7acb859f31"]["object_data"]["bbox"] = [ | ||
json_data["_understand_ai_t4_format/bounding_box_2d_raillabel"] | ||
] | ||
ground_truth["objects"]["48c988bd-76f1-423f-b46d-7e7acb859f31"]["object_data"]["cuboid"] = [ | ||
json_data["_understand_ai_t4_format/bounding_box_3d_raillabel"] | ||
] | ||
ground_truth["objects"]["58e7edd8-a7ee-4775-a837-e6dd375e8150"]["object_data"]["poly2d"] = [ | ||
json_data["_understand_ai_t4_format/polygon_2d_raillabel"] | ||
] | ||
ground_truth["objects"]["4d8eca35-6c1d-4159-8062-21c2f2c051df"]["object_data"]["poly2d"] = [ | ||
json_data["_understand_ai_t4_format/polyline_2d_raillabel"] | ||
] | ||
ground_truth["objects"]["05a7e7a7-91e1-49ef-a172-780f2461f013"]["object_data"]["vec"] = [ | ||
json_data["_understand_ai_t4_format/segmentation_3d_raillabel"] | ||
] | ||
|
||
return ground_truth | ||
|
||
|
||
def test_fromdict(json_data): | ||
input_data = _prepare_frame_data(json_data["_understand_ai_t4_format/frame"], json_data) | ||
frame = uai_format.Frame.fromdict(input_data) | ||
|
||
assert frame.id == int(input_data["frameId"]) | ||
assert frame.timestamp == Decimal(input_data["timestamp"]) | ||
assert frame.bounding_box_2ds[json_data["_understand_ai_t4_format/bounding_box_2d"]["id"]] == uai_format.BoundingBox2d.fromdict( | ||
json_data["_understand_ai_t4_format/bounding_box_2d"] | ||
) | ||
assert frame.polyline_2ds[json_data["_understand_ai_t4_format/polyline_2d"]["id"]] == uai_format.Polyline2d.fromdict( | ||
json_data["_understand_ai_t4_format/polyline_2d"] | ||
# == Fixtures ========================= | ||
|
||
@pytest.fixture | ||
def frame_uai_dict( | ||
bounding_box_2d_uai_dict, | ||
bounding_box_3d_uai_dict, | ||
polygon_2d_uai_dict, | ||
polyline_2d_uai_dict, | ||
segmentation_3d_uai_dict, | ||
sensor_lidar_uai_dict, | ||
) -> dict: | ||
return { | ||
"frameId": "000", | ||
"timestamp": sensor_lidar_uai_dict["timestamp"], | ||
"annotations": { | ||
"2D_BOUNDING_BOX": [bounding_box_2d_uai_dict], | ||
"2D_POLYLINE": [polyline_2d_uai_dict], | ||
"2D_POLYGON": [polygon_2d_uai_dict], | ||
"3D_BOUNDING_BOX": [bounding_box_3d_uai_dict], | ||
"3D_SEGMENTATION": [segmentation_3d_uai_dict], | ||
} | ||
} | ||
|
||
@pytest.fixture | ||
def frame_uai( | ||
bounding_box_2d_uai, | ||
bounding_box_3d_uai, | ||
polygon_2d_uai, | ||
polyline_2d_uai, | ||
segmentation_3d_uai, | ||
sensor_lidar_uai, | ||
): | ||
return uai_format.Frame( | ||
id=0, | ||
timestamp=sensor_lidar_uai.timestamp, | ||
bounding_box_2ds={str(bounding_box_2d_uai.id): bounding_box_2d_uai}, | ||
bounding_box_3ds={str(bounding_box_3d_uai.id): bounding_box_3d_uai}, | ||
polygon_2ds={str(polygon_2d_uai.id): polygon_2d_uai}, | ||
polyline_2ds={str(polyline_2d_uai.id): polyline_2d_uai}, | ||
segmentation_3ds={str(segmentation_3d_uai.id): segmentation_3d_uai}, | ||
) | ||
assert frame.polygon_2ds[json_data["_understand_ai_t4_format/polygon_2d"]["id"]] == uai_format.Polygon2d.fromdict( | ||
json_data["_understand_ai_t4_format/polygon_2d"] | ||
) | ||
assert frame.bounding_box_3ds[json_data["_understand_ai_t4_format/bounding_box_3d"]["id"]] == uai_format.BoundingBox3d.fromdict( | ||
json_data["_understand_ai_t4_format/bounding_box_3d"] | ||
) | ||
assert frame.segmentation_3ds[json_data["_understand_ai_t4_format/segmentation_3d"]["id"]] == uai_format.Segmentation3d.fromdict( | ||
json_data["_understand_ai_t4_format/segmentation_3d"] | ||
) | ||
|
||
def test_to_raillabel(json_data): | ||
input_data = _prepare_frame_data(json_data["_understand_ai_t4_format/frame"], json_data) | ||
frame = uai_format.Frame.fromdict(input_data) | ||
output_data = frame.to_raillabel() | ||
ground_truth = _prepare_ground_truth(json_data["_understand_ai_t4_format/frame_raillabel"], json_data) | ||
|
||
assert output_data == ground_truth | ||
|
||
def test_translated_sensors(json_data): | ||
input_data = _prepare_frame_data(json_data["_understand_ai_t4_format/frame"], json_data) | ||
frame = uai_format.Frame.fromdict(input_data) | ||
ground_truth_streams = json_data["_understand_ai_t4_format/frame_raillabel"]["frame_properties"]["streams"] | ||
|
||
assert frame.translated_sensors.keys() == ground_truth_streams.keys() | ||
|
||
def test_translated_objects(json_data): | ||
input_data = _prepare_frame_data(json_data["_understand_ai_t4_format/frame"], json_data) | ||
frame = uai_format.Frame.fromdict(input_data) | ||
ground_truth_objects = json_data["_understand_ai_t4_format/frame_raillabel"]["objects"] | ||
@pytest.fixture | ||
def frame_raillabel_dict( | ||
bounding_box_2d_uai, bounding_box_2d_raillabel_dict, | ||
bounding_box_3d_raillabel_dict, | ||
polygon_2d_uai, polygon_2d_raillabel_dict, | ||
polyline_2d_uai, polyline_2d_raillabel_dict, | ||
segmentation_3d_uai, segmentation_3d_raillabel_dict, | ||
sensor_lidar_uai, sensor_lidar_raillabel_dict, coordinate_system_lidar_translated_uid, | ||
sensor_camera_raillabel_dict, coordinate_system_camera_translated_uid, | ||
) -> dict: | ||
return { | ||
"frame_properties": { | ||
"timestamp": str(sensor_lidar_uai.timestamp), | ||
"streams": { | ||
coordinate_system_camera_translated_uid: sensor_camera_raillabel_dict, | ||
coordinate_system_lidar_translated_uid: sensor_lidar_raillabel_dict, | ||
} | ||
}, | ||
"objects": { | ||
str(bounding_box_2d_uai.object_id): { | ||
"object_data": { | ||
"bbox": [bounding_box_2d_raillabel_dict], | ||
"cuboid": [bounding_box_3d_raillabel_dict], | ||
} | ||
}, | ||
str(polygon_2d_uai.object_id): { | ||
"object_data": { | ||
"poly2d": [polygon_2d_raillabel_dict] | ||
} | ||
}, | ||
str(polyline_2d_uai.object_id): { | ||
"object_data": { | ||
"poly2d": [polyline_2d_raillabel_dict] | ||
} | ||
}, | ||
str(segmentation_3d_uai.object_id): { | ||
"object_data": { | ||
"vec": [segmentation_3d_raillabel_dict] | ||
} | ||
}, | ||
} | ||
} | ||
|
||
# == Tests ============================ | ||
|
||
def test_fromdict( | ||
bounding_box_2d_uai_dict, bounding_box_2d_uai, | ||
bounding_box_3d_uai_dict, bounding_box_3d_uai, | ||
polygon_2d_uai_dict, polygon_2d_uai, | ||
polyline_2d_uai_dict, polyline_2d_uai, | ||
segmentation_3d_uai_dict, segmentation_3d_uai, | ||
sensor_lidar_uai_dict, sensor_lidar_uai, | ||
): | ||
frame = uai_format.Frame.fromdict( | ||
{ | ||
"frameId": "000", | ||
"timestamp": sensor_lidar_uai_dict["timestamp"], | ||
"annotations": { | ||
"2D_BOUNDING_BOX": [bounding_box_2d_uai_dict], | ||
"2D_POLYLINE": [polyline_2d_uai_dict], | ||
"2D_POLYGON": [polygon_2d_uai_dict], | ||
"3D_BOUNDING_BOX": [bounding_box_3d_uai_dict], | ||
"3D_SEGMENTATION": [segmentation_3d_uai_dict], | ||
} | ||
} | ||
) | ||
|
||
assert frame.translated_objects.keys() == ground_truth_objects.keys() | ||
assert frame.id == 0 | ||
assert frame.timestamp == sensor_lidar_uai.timestamp | ||
assert frame.bounding_box_2ds == {str(bounding_box_2d_uai.id): bounding_box_2d_uai} | ||
assert frame.bounding_box_3ds == {str(bounding_box_3d_uai.id): bounding_box_3d_uai} | ||
assert frame.polygon_2ds == {str(polygon_2d_uai.id): polygon_2d_uai} | ||
assert frame.polyline_2ds == {str(polyline_2d_uai.id): polyline_2d_uai} | ||
assert frame.segmentation_3ds == {str(segmentation_3d_uai.id): segmentation_3d_uai} | ||
|
||
|
||
def test_to_raillabel( | ||
bounding_box_2d_uai, bounding_box_2d_raillabel_dict, | ||
bounding_box_3d_uai, bounding_box_3d_raillabel_dict, | ||
polygon_2d_uai, polygon_2d_raillabel_dict, | ||
polyline_2d_uai, polyline_2d_raillabel_dict, | ||
segmentation_3d_uai, segmentation_3d_raillabel_dict, | ||
sensor_lidar_uai, sensor_lidar_raillabel_dict, coordinate_system_lidar_translated_uid, | ||
sensor_camera_raillabel_dict, coordinate_system_camera_translated_uid, | ||
): | ||
frame = uai_format.Frame( | ||
id=0, | ||
timestamp=sensor_lidar_uai.timestamp, | ||
bounding_box_2ds={str(bounding_box_2d_uai.id): bounding_box_2d_uai}, | ||
bounding_box_3ds={str(bounding_box_3d_uai.id): bounding_box_3d_uai}, | ||
polygon_2ds={str(polygon_2d_uai.id): polygon_2d_uai}, | ||
polyline_2ds={str(polyline_2d_uai.id): polyline_2d_uai}, | ||
segmentation_3ds={str(segmentation_3d_uai.id): segmentation_3d_uai}, | ||
) | ||
|
||
assert frame.to_raillabel() == { | ||
"frame_properties": { | ||
"timestamp": str(sensor_lidar_uai.timestamp), | ||
"streams": { | ||
coordinate_system_camera_translated_uid: sensor_camera_raillabel_dict, | ||
coordinate_system_lidar_translated_uid: sensor_lidar_raillabel_dict, | ||
} | ||
}, | ||
"objects": { | ||
str(bounding_box_2d_uai.object_id): { | ||
"object_data": { | ||
"bbox": [bounding_box_2d_raillabel_dict], | ||
"cuboid": [bounding_box_3d_raillabel_dict], | ||
} | ||
}, | ||
str(polygon_2d_uai.object_id): { | ||
"object_data": { | ||
"poly2d": [polygon_2d_raillabel_dict] | ||
} | ||
}, | ||
str(polyline_2d_uai.object_id): { | ||
"object_data": { | ||
"poly2d": [polyline_2d_raillabel_dict] | ||
} | ||
}, | ||
str(segmentation_3d_uai.object_id): { | ||
"object_data": { | ||
"vec": [segmentation_3d_raillabel_dict] | ||
} | ||
}, | ||
} | ||
} | ||
|
||
# Executes the test if the file is called | ||
if __name__ == "__main__": | ||
import os | ||
os.system("clear") | ||
pytest.main([__file__, "--disable-pytest-warnings", "--cache-clear"]) | ||
pytest.main([__file__, "--disable-pytest-warnings", "--cache-clear", "-vv"]) |
Oops, something went wrong.