Add Cityscapes format (#3758)

CHANGELOG.md

@@ -16,6 +16,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Add a tutorial on attaching cloud storage AWS-S3 (<https://github.com/openvinotoolkit/cvat/pull/3745>)
   and Azure Blob Container (<https://github.com/openvinotoolkit/cvat/pull/3778>)
 - The feature to remove annotations in a specified range of frames (<https://github.com/openvinotoolkit/cvat/pull/3617>)
+- Add Cityscapes format (<https://github.com/openvinotoolkit/cvat/pull/3758>)
 - Add Open Images V6 format (<https://github.com/openvinotoolkit/cvat/pull/3679>)
 
 ### Changed

cvat/apps/dataset_manager/formats/cityscapes.py

+# Copyright (C) 2021 Intel Corporation
+#
+# SPDX-License-Identifier: MIT
+
+import os.path as osp
+from tempfile import TemporaryDirectory
+
+from datumaro.components.dataset import Dataset
+from datumaro.plugins.cityscapes_format import write_label_map
+from pyunpack import Archive
+
+from cvat.apps.dataset_manager.bindings import (GetCVATDataExtractor,
+                                                import_dm_annotations)
+from cvat.apps.dataset_manager.util import make_zip_archive
+
+from .registry import dm_env, exporter, importer
+from .utils import make_colormap
+
+
+@exporter(name='Cityscapes', ext='ZIP', version='1.0')
+def _export(dst_file, instance_data, save_images=False):
+    dataset = Dataset.from_extractors(GetCVATDataExtractor(
+        instance_data, include_images=save_images), env=dm_env)
+    dataset.transform('polygons_to_masks')
+    dataset.transform('boxes_to_masks')
+    dataset.transform('merge_instance_segments')
+    with TemporaryDirectory() as temp_dir:
+        dataset.export(temp_dir, 'cityscapes', save_images=save_images,
+            apply_colormap=True, label_map={label: info[0]
+                for label, info in make_colormap(instance_data).items()})
+
+        make_zip_archive(temp_dir, dst_file)
+
+@importer(name='Cityscapes', ext='ZIP', version='1.0')
+def _import(src_file, instance_data):
+    with TemporaryDirectory() as tmp_dir:
+        Archive(src_file.name).extractall(tmp_dir)
+
+        labelmap_file = osp.join(tmp_dir, 'label_colors.txt')
+        if not osp.isfile(labelmap_file):
+            colormap = {label: info[0]
+                for label, info in make_colormap(instance_data).items()}
+            write_label_map(labelmap_file, colormap)
+
+        dataset = Dataset.import_from(tmp_dir, 'cityscapes', env=dm_env)
+        dataset.transform('masks_to_polygons')
+        import_dm_annotations(dataset, instance_data)

cvat/apps/dataset_manager/formats/registry.py

@@ -121,5 +121,6 @@ import cvat.apps.dataset_manager.formats.market1501
 import cvat.apps.dataset_manager.formats.icdar
 import cvat.apps.dataset_manager.formats.velodynepoint
 import cvat.apps.dataset_manager.formats.pointcloud
+import cvat.apps.dataset_manager.formats.cityscapes
 import cvat.apps.dataset_manager.formats.openimages
 

cvat/apps/dataset_manager/tests/assets/annotations.json

@@ -1197,5 +1197,23 @@
       }
     ],
     "tracks": []
+  },
+  "Cityscapes 1.0": {
+    "version": 0,
+    "tags": [],
+    "shapes": [
+      {
+        "type": "polygon",
+        "occluded": false,
+        "z_order": 0,
+        "points": [11.9, 21.5, 35.2, 21.9, 33.6, 31.9, 12.4, 30.47],
+        "frame": 0,
+        "label_id": null,
+        "group": 0,
+        "source": "manual",
+        "attributes": []
+      }
+    ],
+    "tracks": []
   }
 }

cvat/apps/dataset_manager/tests/assets/tasks.json

@@ -54,7 +54,7 @@
       }
     ]
   },
-  "icdar_localization_and_recognition": {
+  "ICDAR Localization 1.0": {
     "name": "icdar localization/recogntion task",
     "overlap": 0,
     "segment_size": 100,
@@ -74,7 +74,27 @@
       }
     ]
   },
-  "icdar_segmentation": {
+  "ICDAR Recognition 1.0": {
+    "name": "icdar localization/recogntion task",
+    "overlap": 0,
+    "segment_size": 100,
+    "owner_id": 1,
+    "assignee_id": 2,
+    "labels": [
+      {
+        "name": "icdar",
+        "attributes": [
+          {
+            "name": "text",
+            "mutable": false,
+            "input_type": "text",
+            "values": ["word_1", "word_2", "word_3"]
+          }
+        ]
+      }
+    ]
+  },
+  "ICDAR Segmentation 1.0": {
     "name": "icdar segmentation task",
     "overlap": 0,
     "segment_size": 100,
@@ -112,7 +132,7 @@
       }
     ]
   },
-  "market1501": {
+  "Market-1501 1.0": {
     "name": "market1501 task",
     "overlap": 0,
     "segment_size": 100,
@@ -144,6 +164,38 @@
       }
     ]
   },
+  "Cityscapes 1.0": {
+    "name": "cityscapes task",
+    "overlap": 0,
+    "segment_size": 100,
+    "owner_id": 1,
+    "assignee_id": 2,
+    "labels": [
+      {
+        "name": "car",
+        "color": "#2080c0",
+        "attributes": [
+          {
+            "name": "is_crowd",
+            "mutable": false,
+            "input_type": "checkbox",
+            "default_value": "false",
+            "values": ["false", "true"]
+          }
+        ]
+      },
+      {
+        "name": "person",
+        "color": "#c06060",
+        "attributes": []
+      },
+      {
+        "name": "background",
+        "color": "#000000",
+        "attributes": []
+      }
+    ]
+  },
   "wrong_checkbox_value": {
     "name": "wrong checkbox value task",
     "overlap": 0,

cvat/apps/dataset_manager/tests/test_formats.py

@@ -27,7 +27,7 @@ from cvat.apps.dataset_manager.util import make_zip_archive
 from cvat.apps.engine.models import Task
 
 
-def generate_image_file(filename, size=(100, 50)):
+def generate_image_file(filename, size=(100, 100)):
     f = BytesIO()
     image = Image.new('RGB', size=size)
     image.save(f, 'jpeg')
@@ -296,6 +296,7 @@ class TaskExportTest(_DbTestBase):
             'ICDAR Segmentation 1.0',
             'Kitti Raw Format 1.0',
             'Sly Point Cloud Format 1.0',
+            'Cityscapes 1.0',
             'Open Images V6 1.0'
         })
 
@@ -323,6 +324,7 @@ class TaskExportTest(_DbTestBase):
             'ICDAR Segmentation 1.0',
             'Kitti Raw Format 1.0',
             'Sly Point Cloud Format 1.0',
+            'Cityscapes 1.0',
             'Open Images V6 1.0',
             'Datumaro 1.0',
             'Datumaro 3D 1.0'
@@ -372,6 +374,7 @@ class TaskExportTest(_DbTestBase):
             ('ICDAR Recognition 1.0', 'icdar_word_recognition'),
             ('ICDAR Localization 1.0', 'icdar_text_localization'),
             ('ICDAR Segmentation 1.0', 'icdar_text_segmentation'),
+            # ('Cityscapes 1.0', 'cityscapes'), does not support, empty annotations
         ]:
             with self.subTest(format=format_name):
                 if not dm.formats.registry.EXPORT_FORMATS[format_name].ENABLED:
@@ -382,16 +385,6 @@ class TaskExportTest(_DbTestBase):
 
                 def check(file_path):
                     def load_dataset(src):
-                        if importer_name == 'datumaro_project':
-                            project = datumaro.components.project. \
-                                Project.load(src)
-
-                            # NOTE: can't import cvat.utils.cli
-                            # for whatever reason, so remove the dependency
-                            #
-                            project.config.remove('sources')
-
-                            return project.make_dataset()
                         return datumaro.components.dataset. \
                             Dataset.import_from(src, importer_name, env=dm_env)
 
@@ -702,6 +695,10 @@ class TaskAnnotationsImportTest(_DbTestBase):
                         }
                     ]
                 },
+                {
+                    "name": "background",
+                    "attributes": [],
+                },
                 {"name": "person"}
             ]
 
@@ -838,7 +835,7 @@ class TaskAnnotationsImportTest(_DbTestBase):
                     {
                         "frame": 0,
                         "attributes": [],
-                        "points": [1.0, 2.1, 10.6, 53.22, 100, 300.222],
+                        "points": [1.0, 2.1, 10.6, 53.22, 90, 90.222],
                         "type": "polygon",
                         "occluded": False,
                         "outside": False
@@ -879,10 +876,10 @@ class TaskAnnotationsImportTest(_DbTestBase):
             elif annotation_format == "MOTS PNG 1.0":
                 tracks = [track_wo_attrs]
             else:
-                shapes = [rectangle_shape_wo_attrs,
+                shapes = [rectangle_shape_wo_attrs, \
                     rectangle_shape_with_attrs]
-                tags = tag_wo_attrs
-                tracks = track_wo_attrs
+                tags = [tag_wo_attrs]
+                tracks = [track_wo_attrs]
 
         annotations = {
             "version": 0,

cvat/apps/dataset_manager/tests/test_rest_api_formats.py

@@ -355,12 +355,12 @@ class TaskDumpUploadTest(_DbTestBase):
                 with self.subTest(format=dump_format_name):
                     images = self._generate_task_images(3)
                     # create task with annotations
-                    if dump_format_name == "Market-1501 1.0":
-                        task = self._create_task(tasks["market1501"], images)
-                    elif dump_format_name in ["ICDAR Localization 1.0", "ICDAR Recognition 1.0"]:
-                        task = self._create_task(tasks["icdar_localization_and_recognition"], images)
-                    elif dump_format_name == "ICDAR Segmentation 1.0":
-                        task = self._create_task(tasks["icdar_segmentation"], images)
+                    if dump_format_name in [
+                        "Market-1501 1.0", "Cityscapes 1.0", \
+                        "ICDAR Localization 1.0", "ICDAR Recognition 1.0", \
+                        "ICDAR Segmentation 1.0"
+                    ]:
+                        task = self._create_task(tasks[dump_format_name], images)
                     else:
                         task = self._create_task(tasks["main"], images)
                     task_id = task["id"]
@@ -368,7 +368,8 @@ class TaskDumpUploadTest(_DbTestBase):
                         "MOT 1.1", "MOTS PNG 1.0", \
                         "PASCAL VOC 1.1", "Segmentation mask 1.1", \
                         "TFRecord 1.0", "YOLO 1.1", "ImageNet 1.0", \
-                        "WiderFace 1.0", "VGGFace2 1.0", "Datumaro 1.0"\
+                        "WiderFace 1.0", "VGGFace2 1.0", "Cityscapes 1.0", \
+                        "Datumaro 1.0"\
                     ]:
                         self._create_annotations(task, dump_format_name, "default")
                     else:
@@ -417,12 +418,12 @@ class TaskDumpUploadTest(_DbTestBase):
                         for user, edata in list(expected.items()):
                             # remove all annotations from task (create new task without annotation)
                             images = self._generate_task_images(3)
-                            if upload_format_name == "Market-1501 1.0":
-                                task = self._create_task(tasks["market1501"], images)
-                            elif upload_format_name in ["ICDAR Localization 1.0", "ICDAR Recognition 1.0"]:
-                                task = self._create_task(tasks["icdar_localization_and_recognition"], images)
-                            elif upload_format_name == "ICDAR Segmentation 1.0":
-                                task = self._create_task(tasks["icdar_segmentation"], images)
+                            if upload_format_name in [
+                                "Market-1501 1.0", "Cityscapes 1.0", \
+                                "ICDAR Localization 1.0", "ICDAR Recognition 1.0", \
+                                "ICDAR Segmentation 1.0"
+                            ]:
+                                task = self._create_task(tasks[upload_format_name], images)
                             else:
                                 task = self._create_task(tasks["main"], images)
                             task_id = task["id"]
@@ -460,12 +461,12 @@ class TaskDumpUploadTest(_DbTestBase):
                 with self.subTest(format=dump_format_name):
                     # create task with annotations
                     video = self._generate_task_videos(1)
-                    if dump_format_name == "Market-1501 1.0":
-                        task = self._create_task(tasks["market1501"], video)
-                    elif dump_format_name in ["ICDAR Localization 1.0", "ICDAR Recognition 1.0"]:
-                        task = self._create_task(tasks["icdar_localization_and_recognition"], video)
-                    elif dump_format_name == "ICDAR Segmentation 1.0":
-                        task = self._create_task(tasks["icdar_segmentation"], video)
+                    if dump_format_name in [
+                        "Market-1501 1.0", "Cityscapes 1.0", \
+                        "ICDAR Localization 1.0", "ICDAR Recognition 1.0", \
+                        "ICDAR Segmentation 1.0"
+                    ]:
+                        task = self._create_task(tasks[dump_format_name], video)
                     else:
                         task = self._create_task(tasks["main"], video)
                     task_id = task["id"]
@@ -474,7 +475,7 @@ class TaskDumpUploadTest(_DbTestBase):
                             "MOT 1.1", "MOTS PNG 1.0", \
                             "PASCAL VOC 1.1", "Segmentation mask 1.1", \
                             "TFRecord 1.0", "YOLO 1.1", "ImageNet 1.0", \
-                            "WiderFace 1.0", "VGGFace2 1.0", \
+                            "WiderFace 1.0", "VGGFace2 1.0", "Cityscapes 1.0" \
                     ]:
                         self._create_annotations(task, dump_format_name, "default")
                     else:
@@ -521,12 +522,12 @@ class TaskDumpUploadTest(_DbTestBase):
                         for user, edata in list(expected.items()):
                             # remove all annotations from task (create new task without annotation)
                             video = self._generate_task_videos(1)
-                            if upload_format_name == "Market-1501 1.0":
-                                task = self._create_task(tasks["market1501"], video)
-                            elif upload_format_name in ["ICDAR Localization 1.0", "ICDAR Recognition 1.0"]:
-                                task = self._create_task(tasks["icdar_localization_and_recognition"], video)
-                            elif upload_format_name == "ICDAR Segmentation 1.0":
-                                task = self._create_task(tasks["icdar_segmentation"], video)
+                            if upload_format_name in [
+                                "Market-1501 1.0", "Cityscapes 1.0", \
+                                "ICDAR Localization 1.0", "ICDAR Recognition 1.0", \
+                                "ICDAR Segmentation 1.0"
+                            ]:
+                                task = self._create_task(tasks[upload_format_name], video)
                             else:
                                 task = self._create_task(tasks["main"], video)
                             task_id = task["id"]
@@ -802,12 +803,12 @@ class TaskDumpUploadTest(_DbTestBase):
                 with self.subTest(format=dump_format_name):
                     images = self._generate_task_images(3)
                     # create task with annotations
-                    if dump_format_name == "Market-1501 1.0":
-                        task = self._create_task(tasks["market1501"], images)
-                    elif dump_format_name in ["ICDAR Localization 1.0", "ICDAR Recognition 1.0"]:
-                        task = self._create_task(tasks["icdar_localization_and_recognition"], images)
-                    elif dump_format_name == "ICDAR Segmentation 1.0":
-                        task = self._create_task(tasks["icdar_segmentation"], images)
+                    if dump_format_name in [
+                        "Market-1501 1.0", "Cityscapes 1.0", \
+                        "ICDAR Localization 1.0", "ICDAR Recognition 1.0", \
+                        "ICDAR Segmentation 1.0"
+                    ]:
+                        task = self._create_task(tasks[dump_format_name], images)
                     else:
                         task = self._create_task(tasks["main"], images)
                     task_id = task["id"]
@@ -871,6 +872,7 @@ class TaskDumpUploadTest(_DbTestBase):
                 with self.subTest(format=upload_format_name):
                     if upload_format_name in [
                         "MOTS PNG 1.0",  # issue #2925 and changed points values
+                        "Cityscapes 1.0" # formats doesn't support empty annotations
                     ]:
                         self.skipTest("Format is fail")
                     images = self._generate_task_images(3)
@@ -899,25 +901,26 @@ class TaskDumpUploadTest(_DbTestBase):
                         "MOTS PNG 1.0",  # issue #2925 and changed points values
                         'Kitti Raw Format 1.0',
                         'Sly Point Cloud Format 1.0',
-                        'Datumaro 3D 1.0'
+                        'Datumaro 3D 1.0',
+                        "Cityscapes 1.0" # expanding annotations due to background mask
                     ]:
                         self.skipTest("Format is fail")
                     images = self._generate_task_images(3)
-                    if dump_format_name == "Market-1501 1.0":
-                        task = self._create_task(tasks["market1501"], images)
-                    elif dump_format_name in ["ICDAR Localization 1.0", "ICDAR Recognition 1.0"]:
-                        task = self._create_task(tasks["icdar_localization_and_recognition"], images)
-                    elif dump_format_name == "ICDAR Segmentation 1.0":
-                        task = self._create_task(tasks["icdar_segmentation"], images)
+                    if dump_format_name in [
+                        "Market-1501 1.0", "Cityscapes 1.0", \
+                        "ICDAR Localization 1.0", "ICDAR Recognition 1.0", \
+                        "ICDAR Segmentation 1.0"
+                    ]:
+                        task = self._create_task(tasks[dump_format_name], images)
                     else:
                         task = self._create_task(tasks["main"], images)
                     task_id = task["id"]
                     if dump_format_name in [
-                        "MOT 1.1", "MOTS PNG 1.0", \
-                        "PASCAL VOC 1.1", "Segmentation mask 1.1", \
-                        "TFRecord 1.0", "YOLO 1.1", "ImageNet 1.0", \
-                        "WiderFace 1.0", "VGGFace2 1.0", "Datumaro 1.0",\
-                        "Open Images V6 1.0" \
+                        "MOT 1.1", "MOTS PNG 1.0",
+                        "PASCAL VOC 1.1", "Segmentation mask 1.1",
+                        "TFRecord 1.0", "YOLO 1.1", "ImageNet 1.0",
+                        "WiderFace 1.0", "VGGFace2 1.0", "Cityscapes 1.0",
+                        "Datumaro 1.0", "Open Images V6 1.0"
                     ]:
                         self._create_annotations(task, dump_format_name, "default")
                     else:
@@ -1009,19 +1012,19 @@ class TaskDumpUploadTest(_DbTestBase):
                         "Open Images V6 1.0", # changed points values
                         'Kitti Raw Format 1.0',
                         'Sly Point Cloud Format 1.0',
+                        'Cityscapes 1.0', # changed points value
                         'Datumaro 3D 1.0'
                     ]:
                         self.skipTest("Format is fail")
 
                     # create task
                     images = self._generate_task_images(3)
-                    if dump_format_name == "Market-1501 1.0":
-                        task = self._create_task(tasks["market1501"], images)
-                    elif dump_format_name in ["ICDAR Localization 1.0",
-                            "ICDAR Recognition 1.0"]:
-                        task = self._create_task(tasks["icdar_localization_and_recognition"], images)
-                    elif dump_format_name == "ICDAR Segmentation 1.0":
-                        task = self._create_task(tasks["icdar_segmentation"], images)
+                    if dump_format_name in [
+                        "Market-1501 1.0", "Cityscapes 1.0", \
+                        "ICDAR Localization 1.0", "ICDAR Recognition 1.0", \
+                        "ICDAR Segmentation 1.0"
+                    ]:
+                        task = self._create_task(tasks[dump_format_name], images)
                     else:
                         task = self._create_task(tasks["main"], images)
 

cvat/apps/engine/tests/test_rest_api.py

@@ -4810,7 +4810,9 @@ class TaskAnnotationAPITestCase(JobAnnotationAPITestCase):
                 annotations["tags"] = tags_wo_attrs
                 annotations["shapes"] = points_wo_attrs \
                                       + rectangle_shapes_wo_attrs
-
+            elif annotation_format == "Cityscapes 1.0":
+                annotations["shapes"] = points_wo_attrs \
+                                      + rectangle_shapes_wo_attrs
             elif annotation_format == "Open Images V6 1.0":
                 annotations["tags"] = tags_wo_attrs
                 annotations["shapes"] = rectangle_shapes_wo_attrs \

site/content/en/docs/manual/advanced/formats/format-cityscapes.md

+---
+linkTitle: 'Cityscapes'
+weight: 16
+---
+
+# [Cityscapes](https://www.cityscapes-dataset.com/login/)
+
+- [Format specification](https://github.com/mcordts/cityscapesScripts#the-cityscapes-dataset)
+
+- Supported annotations
+
+  - Polygons (segmentation task)
+
+- Supported attributes
+  - 'is_crowd' (boolean, should be defined for labels as `checkbox` -es)
+    Specifies if the annotation label can distinguish between different instances.
+    If False, the annotation id field encodes the instance id.
+
+# Cityscapes export
+
+Downloaded file: a zip archive of the following structure:
+
+```
+.
+├── label_color.txt
+├── gtFine
+│   ├── <subset_name>
+│   │   └── <city_name>
+│   │       ├── image_0_gtFine_instanceIds.png
+│   │       ├── image_0_gtFine_color.png
+│   │       ├── image_0_gtFine_labelIds.png
+│   │       ├── image_1_gtFine_instanceIds.png
+│   │       ├── image_1_gtFine_color.png
+│   │       ├── image_1_gtFine_labelIds.png
+│   │       ├── ...
+└── imgsFine  # if saving images was requested
+    └── leftImg8bit
+        ├── <subset_name>
+        │   └── <city_name>
+        │       ├── image_0_leftImg8bit.png
+        │       ├── image_1_leftImg8bit.png
+        │       ├── ...
+```
+
+- `label_color.txt` a file that describes the color for each label
+
+```
+# label_color.txt example
+# r g b label_name
+0 0 0 background
+0 255 0 tree
+...
+```
+
+- `*_gtFine_color.png` class labels encoded by its color.
+- `*_gtFine_labelIds.png` class labels are encoded by its index.
+- `*_gtFine_instanceIds.png` class and instance labels encoded
+  by an instance ID. The pixel values encode class and the individual instance:
+  the integer part of a division by 1000 of each ID provides class ID,
+  the remainder is the instance ID. If a certain annotation describes multiple
+  instances, then the pixels have the regular ID of that class
+
+# Cityscapes annotations import
+
+Uploaded file: a zip archive with the following structure:
+
+```
+.
+├── label_color.txt # optional
+└── gtFine
+    └── <city_name>
+        ├── image_0_gtFine_instanceIds.png
+        ├── image_1_gtFine_instanceIds.png
+        ├── ...
+```
+
+# Creating task with Cityscapes dataset
+
+Create a task with the labels you need
+or you can use the labels and colors of the original dataset.
+To work with the Cityscapes format, you must have a black color label
+for the background.
+
+Original Cityscapes color map:
+
+<details>
+
+```JSON
+[
+    {"name": "unlabeled", "color": "#000000", "attributes": []},
+    {"name": "egovehicle", "color": "#000000", "attributes": []},
+    {"name": "rectificationborder", "color": "#000000", "attributes": []},
+    {"name": "outofroi", "color": "#000000", "attributes": []},
+    {"name": "static", "color": "#000000", "attributes": []},
+    {"name": "dynamic", "color": "#6f4a00", "attributes": []},
+    {"name": "ground", "color": "#510051", "attributes": []},
+    {"name": "road", "color": "#804080", "attributes": []},
+    {"name": "sidewalk", "color": "#f423e8", "attributes": []},
+    {"name": "parking", "color": "#faaaa0", "attributes": []},
+    {"name": "railtrack", "color": "#e6968c", "attributes": []},
+    {"name": "building", "color": "#464646", "attributes": []},
+    {"name": "wall", "color": "#66669c", "attributes": []},
+    {"name": "fence", "color": "#be9999", "attributes": []},
+    {"name": "guardrail", "color": "#b4a5b4", "attributes": []},
+    {"name": "bridge", "color": "#966464", "attributes": []},
+    {"name": "tunnel", "color": "#96785a", "attributes": []},
+    {"name": "pole", "color": "#999999", "attributes": []},
+    {"name": "polegroup", "color": "#999999", "attributes": []},
+    {"name": "trafficlight", "color": "#faaa1e", "attributes": []},
+    {"name": "trafficsign", "color": "#dcdc00", "attributes": []},
+    {"name": "vegetation", "color": "#6b8e23", "attributes": []},
+    {"name": "terrain", "color": "#98fb98", "attributes": []},
+    {"name": "sky", "color": "#4682b4", "attributes": []},
+    {"name": "person", "color": "#dc143c", "attributes": []},
+    {"name": "rider", "color": "#ff0000", "attributes": []},
+    {"name": "car", "color": "#00008e", "attributes": []},
+    {"name": "truck", "color": "#000046", "attributes": []},
+    {"name": "bus", "color": "#003c64", "attributes": []},
+    {"name": "caravan", "color": "#00005a", "attributes": []},
+    {"name": "trailer", "color": "#00006e", "attributes": []},
+    {"name": "train", "color": "#005064", "attributes": []},
+    {"name": "motorcycle", "color": "#0000e6", "attributes": []},
+    {"name": "bicycle", "color": "#770b20", "attributes": []},
+    {"name": "licenseplate", "color": "#00000e", "attributes": []}
+]
+
+```
+
+<details>
+
+Upload images when creating a task:
+
+```
+images.zip/
+    ├── image_0.jpg
+    ├── image_1.jpg
+    ├── ...
+
+```
+
+After creating the task, upload the Cityscapes annotations as described
+in the previous section.