-
Notifications
You must be signed in to change notification settings - Fork 0
/
photoneo.py
240 lines (190 loc) · 9.54 KB
/
photoneo.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
# Copyright (C) 2024 CVAT.ai Corporation
#
# SPDX-License-Identifier: MIT
import zipfile
from datumaro.components.dataset import Dataset
from datumaro.components.annotation import AnnotationType, LabelCategories
from datumaro.plugins.coco_format.importer import CocoImporter
from datumaro.components.annotation import Points
from cvat.apps.dataset_manager.bindings import GetCVATDataExtractor, detect_dataset, \
import_dm_annotations
from cvat.apps.dataset_manager.util import make_zip_archive
from .registry import dm_env, exporter, importer
# my own logger config for debugging
import logging
import json
import os
# import debugpy
my_export_logger = logging.getLogger("my_export_logger")
my_export_logger.setLevel(logging.DEBUG)
my_fh = logging.FileHandler('/home/django/logs/my_log.log', encoding='utf-8')
my_fh.setLevel(logging.DEBUG)
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
my_fh.setFormatter(formatter)
my_export_logger.addHandler(my_fh)
my_export_logger.info('first')
@importer(name="COCO Photoneo", version="1.0", ext="ZIP")
def my_importer(src_file, temp_dir, instance_data, load_data_callback=None, **kwargs):
if zipfile.is_zipfile(src_file):
zipfile.ZipFile(src_file).extractall(temp_dir)
# We use coco importer because it gives better error message
detect_dataset(temp_dir, format_name='coco', importer=CocoImporter)
dataset = Dataset.import_from(temp_dir, 'coco_instances', env=dm_env)
if load_data_callback is not None:
load_data_callback(dataset, instance_data)
json_path = temp_dir+"/annotations/instances_default.json"
with open(json_path, 'r') as file:
data = json.load(file)
label_categories = dataset.categories().get(AnnotationType.label, None)
for category in data["categories"]:
for kp in category["keypoints"]:
label_categories.add(category["name"]+kp)
dataset.categories()[AnnotationType.label] = label_categories
my_export_logger.info(f"dataset_categories: {dataset.categories().get(AnnotationType.label, None)}")
for ann in data["annotations"]:
file_name = find_property_by_id(data["images"], ann["image_id"], "file_name")
keypoints = find_property_by_id(data["categories"], ann["category_id"], "keypoints")
ann_label = find_property_by_id(data["categories"], ann["category_id"], "name")
my_export_logger.info(f"ann : {ann_label}, {keypoints}")
my_export_logger.info(f"keypoints: {ann}")
for i, kp in enumerate(keypoints):
x = ann["keypoints"][i*3]
y = ann["keypoints"][i*3+1]
vis = ann["keypoints"][i*3+2]
if vis:
item = dataset.get(file_name.split(".")[0])
label_index = label_categories.find(ann_label+kp)
my_export_logger.info(f"label_index: {label_index[0]}")
item.annotations.append(Points([x,y], label=label_index[0]))
my_export_logger.info(f"item: {item}")
dataset.put(item)
# my_export_logger.info(f"new item: {dataset.get(file_name.split(".")[0])}")
import_dm_annotations(dataset, instance_data)
else:
dataset = Dataset.import_from(src_file.name,
'coco_instances', env=dm_env)
import_dm_annotations(dataset, instance_data)
def get_image_id(images, file_name):
for image in images:
if image["file_name"] == file_name:
return image["id"]
return None
def bbox_to_flat_polygon(bbox):
x, y, w, h = bbox
return [x, y, x+w, y, x+w, y+h, x, y+h]
def is_point_in_polygon(point, flat_polygon):
"""
Check if a point is inside a polygon using the ray-casting algorithm.
Args:
point: A tuple (x, y) representing the coordinates of the point.
polygon: A list of tuples [(x1, y1), (x2, y2), ..., (xn, yn)] representing the vertices of the polygon.
Returns:
True if the point is inside the polygon, False otherwise.
"""
polygon = [(flat_polygon[i], flat_polygon[i + 1]) for i in range(0, len(flat_polygon), 2)]
x, y = point
num_vertices = len(polygon)
inside = False
p1x, p1y = polygon[0]
for i in range(num_vertices + 1):
p2x, p2y = polygon[i % num_vertices]
if y > min(p1y, p2y):
if y <= max(p1y, p2y):
if x <= max(p1x, p2x):
if p1y != p2y:
xinters = (y - p1y) * (p2x - p1x) / (p2y - p1y) + p1x
if p1x == p2x or x <= xinters:
inside = not inside
p1x, p1y = p2x, p2y
# my_export_logger.info('testing kp: {}, if it is inside flat polygon: {}'.format(str(point), str(flat_polygon)))
# if inside:
# my_export_logger.info('jupiii! Is inside flat polygon')
return inside
def create_categories(data):
return data
def find_prefix(string, prefixes):
for prefix in prefixes:
if string.startswith(prefix):
return prefix
return None
def find_property_by_id(categories, id_to_find, property):
for category in categories:
if category['id'] == id_to_find:
return category[property]
return None
def remove_prefix(string, prefix):
if string.startswith(prefix):
return string[len(prefix):]
return string
def insert_keypoint_triplet(data, i, idx, kp):
data["annotations"][i]["num_keypoints"] += 1
# insert a keypoint to correct triple position in 'keypoints': (x, y, visibility)
data["annotations"][i]["keypoints"][3*idx] = kp["kp"][0]
data["annotations"][i]["keypoints"][3*idx + 1] = kp["kp"][1]
data["annotations"][i]["keypoints"][3*idx +2 ] = 2
return data
@exporter(name="COCO Photoneo", version="1.0", ext="ZIP")
def my_exporter(dst_file, temp_dir, instance_data, save_images=False):
with GetCVATDataExtractor(instance_data, include_images=save_images) as extractor:
dataset = Dataset.from_extractors(extractor, env=dm_env)
dataset.export(temp_dir, 'coco_instances', save_images=save_images,
merge_images=True)
my_export_logger.info('dataset: \n{}'.format(dataset.categories().get(AnnotationType.label, None)))
my_export_logger.info(f'tmp: \n{temp_dir}')
json_path = temp_dir+"/annotations/"#instances_default.json"
json_files = []
for fn in os.listdir(json_path):
json_files.append(json_path+fn)
my_export_logger.info(f'file ann: \n{json_files[-1]}')
with open(json_files[-1], 'r') as file:
data = json.load(file)
# collect all categories labels
# create categories object
categories = {}
categories_labels = []
my_export_logger.info(f'data loaded\n')
for shape in instance_data.shapes:
if shape.type == "rectangle" or shape.type == "polygon":
if shape.label not in categories:
# cvat indexes from 1
categories[shape.label] = {"id": len(categories)+1, "name": shape.label, "keypoints": set()}
categories_labels.append(shape.label)
my_export_logger.info(f'shape.label: \n{shape.label}')
categories_labels = sorted(categories_labels, key=len, reverse=True)
points_data = {key: [] for key in categories_labels}
my_export_logger.info('categories: {}'.format(categories))
# extract all keypoints, assign frame name and label to them
for frame_data in instance_data.group_by_frame(include_empty=True):
for shape in frame_data.labeled_shapes:
my_export_logger.info('shape: {}'.format(shape))
if shape.type == "points":
prefix = find_prefix(shape.label, categories_labels)
my_export_logger.info('prefix: {}'.format(str(prefix)))
l = remove_prefix(shape.label, prefix)
points_data[prefix].append({"kp": shape.points, "label":shape.label, "image_id": get_image_id(data["images"], frame_data.name)})
categories[prefix]["keypoints"].add(l)
# format keypoints categories to sorted list
for label_key in categories.keys():
categories[label_key]["keypoints"] = list(sorted(categories[label_key]["keypoints"]))
# reindex annotations to new categories
# create 'num_keypoints' and 'keypoints' fields
for i, ann in enumerate(data["annotations"]):
label = find_property_by_id(data["categories"], ann["category_id"], "name")
data["annotations"][i]["category_id"] = categories[label]["id"]
data["annotations"][i]["num_keypoints"] = 0
data["annotations"][i]["keypoints"] = [0] * (3 * len(categories[label]["keypoints"]))
data["categories"] = list(categories.values())
for i, ann in enumerate(data["annotations"]):
label = find_property_by_id(data["categories"], ann["category_id"], "name")
for kp in points_data[label]:
my_export_logger.info('keypoints: {}'.format(kp))
if ann["image_id"] == kp["image_id"]:
idx = categories[label]["keypoints"].index(remove_prefix(kp["label"], label))
if ann["segmentation"] and is_point_in_polygon(kp["kp"], ann["segmentation"][0]):
data = insert_keypoint_triplet(data, i, idx, kp)
elif ann["bbox"] and is_point_in_polygon(kp["kp"], bbox_to_flat_polygon(ann["bbox"])):
data = insert_keypoint_triplet(data, i, idx, kp)
my_export_logger.info('json data: \n{}'.format(json.dumps(data, indent=4)))
with open(json_files[-1], 'w') as file:
json.dump(data, file)
make_zip_archive(temp_dir, dst_file)