def get_bbox(self) -> Tuple[int, int, int, int]:
"""
Computes the bounding box of the mask.
Returns: [x, y, w, h]
"""
from datumaro.util.mask_tools import find_mask_bbox
return find_mask_bbox(self.image)
def _save_item(self, item, subset_dir, index):
from lxml import etree as ET
log.debug("Converting item '%s'", item.id)
image_filename = self._make_image_filename(item)
if self._save_images:
if item.has_image and item.image.has_data:
self._save_image(item, osp.join(subset_dir, image_filename))
else:
log.debug("Item '%s' has no image", item.id)
root_elem = ET.Element('annotation')
ET.SubElement(root_elem, 'filename').text = osp.basename(image_filename)
ET.SubElement(root_elem, 'folder').text = osp.dirname(image_filename)
source_elem = ET.SubElement(root_elem, 'source')
ET.SubElement(source_elem, 'sourceImage').text = ''
ET.SubElement(source_elem, 'sourceAnnotation').text = 'Datumaro'
if item.has_image:
image_elem = ET.SubElement(root_elem, 'imagesize')
image_size = item.image.size
ET.SubElement(image_elem, 'nrows').text = str(image_size[0])
ET.SubElement(image_elem, 'ncols').text = str(image_size[1])
groups = defaultdict(list)
obj_id = 0
for ann in item.annotations:
if not ann.type in { AnnotationType.polygon,
AnnotationType.bbox, AnnotationType.mask }:
continue
obj_elem = ET.SubElement(root_elem, 'object')
ET.SubElement(obj_elem, 'name').text = self._get_label(ann.label)
ET.SubElement(obj_elem, 'deleted').text = '0'
ET.SubElement(obj_elem, 'verified').text = '0'
ET.SubElement(obj_elem, 'occluded').text = \
'yes' if ann.attributes.pop('occluded', '') == True else 'no'
ET.SubElement(obj_elem, 'date').text = ''
ET.SubElement(obj_elem, 'id').text = str(obj_id)
parts_elem = ET.SubElement(obj_elem, 'parts')
if ann.group:
groups[ann.group].append((obj_id, parts_elem))
else:
ET.SubElement(parts_elem, 'hasparts').text = ''
ET.SubElement(parts_elem, 'ispartof').text = ''
if ann.type == AnnotationType.bbox:
ET.SubElement(obj_elem, 'type').text = 'bounding_box'
poly_elem = ET.SubElement(obj_elem, 'polygon')
x0, y0, x1, y1 = ann.points
points = [ (x0, y0), (x1, y0), (x1, y1), (x0, y1) ]
for x, y in points:
point_elem = ET.SubElement(poly_elem, 'pt')
ET.SubElement(point_elem, 'x').text = '%.2f' % x
ET.SubElement(point_elem, 'y').text = '%.2f' % y
ET.SubElement(poly_elem, 'username').text = \
str(ann.attributes.pop('username', ''))
elif ann.type == AnnotationType.polygon:
poly_elem = ET.SubElement(obj_elem, 'polygon')
for x, y in zip(ann.points[::2], ann.points[1::2]):
point_elem = ET.SubElement(poly_elem, 'pt')
ET.SubElement(point_elem, 'x').text = '%.2f' % x
ET.SubElement(point_elem, 'y').text = '%.2f' % y
ET.SubElement(poly_elem, 'username').text = \
str(ann.attributes.pop('username', ''))
elif ann.type == AnnotationType.mask:
mask_filename = '%s_mask_%s.png' % \
(item.id.replace('/', '_'), obj_id)
save_image(osp.join(subset_dir, LabelMePath.MASKS_DIR,
mask_filename),
self._paint_mask(ann.image))
segm_elem = ET.SubElement(obj_elem, 'segm')
ET.SubElement(segm_elem, 'mask').text = mask_filename
bbox = find_mask_bbox(ann.image)
box_elem = ET.SubElement(segm_elem, 'box')
ET.SubElement(box_elem, 'xmin').text = '%.2f' % bbox[0]
ET.SubElement(box_elem, 'ymin').text = '%.2f' % bbox[1]
ET.SubElement(box_elem, 'xmax').text = \
'%.2f' % (bbox[0] + bbox[2])
ET.SubElement(box_elem, 'ymax').text = \
'%.2f' % (bbox[1] + bbox[3])
ET.SubElement(segm_elem, 'username').text = \
str(ann.attributes.pop('username', ''))
else:
raise NotImplementedError("Unknown shape type '%s'" % ann.type)
attrs = []
for k, v in ann.attributes.items():
attrs.append('%s=%s' % (k, v))
ET.SubElement(obj_elem, 'attributes').text = ', '.join(attrs)
obj_id += 1
for _, group in groups.items():
leader_id, leader_parts_elem = group[0]
leader_parts = [str(o_id) for o_id, _ in group[1:]]
ET.SubElement(leader_parts_elem, 'hasparts').text = \
','.join(leader_parts)
ET.SubElement(leader_parts_elem, 'ispartof').text = ''
for obj_id, parts_elem in group[1:]:
ET.SubElement(parts_elem, 'hasparts').text = ''
ET.SubElement(parts_elem, 'ispartof').text = str(leader_id)
xml_path = osp.join(subset_dir, 'item_%09d.xml' % index)
with open(xml_path, 'w', encoding='utf-8') as f:
xml_data = ET.tostring(root_elem, encoding='unicode',
pretty_print=True)
f.write(xml_data)
def get_bbox(self):
from datumaro.util.mask_tools import find_mask_bbox
return find_mask_bbox(self.image)
def _save_item(self, item, subset_dir):
# Disable B410: import_lxml - the library is used for writing here
from lxml import etree as ET # nosec
log.debug("Converting item '%s'", item.id)
image_filename = self._make_image_filename(item)
if self._save_images:
if item.has_image and item.image.has_data:
self._save_image(item, osp.join(subset_dir, image_filename))
else:
log.debug("Item '%s' has no image", item.id)
root_elem = ET.Element('annotation')
ET.SubElement(root_elem, 'filename').text = osp.basename(image_filename)
ET.SubElement(root_elem, 'folder').text = osp.dirname(image_filename)
source_elem = ET.SubElement(root_elem, 'source')
ET.SubElement(source_elem, 'sourceImage').text = ''
ET.SubElement(source_elem, 'sourceAnnotation').text = 'Datumaro'
if item.has_image:
image_elem = ET.SubElement(root_elem, 'imagesize')
image_size = item.image.size
ET.SubElement(image_elem, 'nrows').text = str(image_size[0])
ET.SubElement(image_elem, 'ncols').text = str(image_size[1])
groups = defaultdict(list)
obj_id = 0
for ann in item.annotations:
if not ann.type in { AnnotationType.polygon,
AnnotationType.bbox, AnnotationType.mask }:
continue
obj_elem = ET.SubElement(root_elem, 'object')
ET.SubElement(obj_elem, 'name').text = self._get_label(ann.label)
ET.SubElement(obj_elem, 'deleted').text = '0'
ET.SubElement(obj_elem, 'verified').text = '0'
ET.SubElement(obj_elem, 'occluded').text = \
'yes' if ann.attributes.get('occluded') is True else 'no'
ET.SubElement(obj_elem, 'date').text = ''
ET.SubElement(obj_elem, 'id').text = str(obj_id)
parts_elem = ET.SubElement(obj_elem, 'parts')
if ann.group:
groups[ann.group].append((obj_id, parts_elem))
else:
ET.SubElement(parts_elem, 'hasparts').text = ''
ET.SubElement(parts_elem, 'ispartof').text = ''
if ann.type == AnnotationType.bbox:
ET.SubElement(obj_elem, 'type').text = 'bounding_box'
poly_elem = ET.SubElement(obj_elem, 'polygon')
x0, y0, x1, y1 = ann.points
points = [ (x0, y0), (x1, y0), (x1, y1), (x0, y1) ]
for x, y in points:
point_elem = ET.SubElement(poly_elem, 'pt')
ET.SubElement(point_elem, 'x').text = '%.2f' % x
ET.SubElement(point_elem, 'y').text = '%.2f' % y
ET.SubElement(poly_elem, 'username').text = \
str(ann.attributes.get('username', ''))
elif ann.type == AnnotationType.polygon:
poly_elem = ET.SubElement(obj_elem, 'polygon')
for x, y in zip(ann.points[::2], ann.points[1::2]):
point_elem = ET.SubElement(poly_elem, 'pt')
ET.SubElement(point_elem, 'x').text = '%.2f' % x
ET.SubElement(point_elem, 'y').text = '%.2f' % y
ET.SubElement(poly_elem, 'username').text = \
str(ann.attributes.get('username', ''))
elif ann.type == AnnotationType.mask:
mask_filename = '%s_mask_%s.png' % (item.id, obj_id)
save_image(osp.join(subset_dir, LabelMePath.MASKS_DIR,
mask_filename),
self._paint_mask(ann.image), create_dir=True)
segm_elem = ET.SubElement(obj_elem, 'segm')
ET.SubElement(segm_elem, 'mask').text = mask_filename
bbox = find_mask_bbox(ann.image)
box_elem = ET.SubElement(segm_elem, 'box')
ET.SubElement(box_elem, 'xmin').text = '%.2f' % bbox[0]
ET.SubElement(box_elem, 'ymin').text = '%.2f' % bbox[1]
ET.SubElement(box_elem, 'xmax').text = \
'%.2f' % (bbox[0] + bbox[2])
ET.SubElement(box_elem, 'ymax').text = \
'%.2f' % (bbox[1] + bbox[3])
ET.SubElement(segm_elem, 'username').text = \
str(ann.attributes.get('username', ''))
else:
raise NotImplementedError("Unknown shape type '%s'" % ann.type)
attrs = []
for k, v in ann.attributes.items():
if k in { 'username' , 'occluded' }:
continue
if isinstance(v, str):
if cast(v, float) is not None and str(float(v)) == v or \
cast(v, int) is not None and str(int(v)) == v:
v = f'"{v}"' # add escaping for string values
else:
v = self._escape(v) # pylint: disable=redundant-keyword-arg due FP https://github.com/PyCQA/pylint/issues/2271
attrs.append('%s=%s' % (self._escape(k), v)) # pylint: disable=redundant-keyword-arg due FP https://github.com/PyCQA/pylint/issues/2271
ET.SubElement(obj_elem, 'attributes').text = ', '.join(attrs)
obj_id += 1
for _, group in groups.items():
leader_id, leader_parts_elem = group[0]
leader_parts = [str(o_id) for o_id, _ in group[1:]]
ET.SubElement(leader_parts_elem, 'hasparts').text = \
','.join(leader_parts)
ET.SubElement(leader_parts_elem, 'ispartof').text = ''
for obj_id, parts_elem in group[1:]:
ET.SubElement(parts_elem, 'hasparts').text = ''
ET.SubElement(parts_elem, 'ispartof').text = str(leader_id)
os.makedirs(osp.join(subset_dir, osp.dirname(image_filename)),
exist_ok=True)
xml_path = osp.join(subset_dir, osp.splitext(image_filename)[0] + '.xml')
if osp.exists(xml_path):
xml_path = osp.join(subset_dir, image_filename + '.xml')
with open(xml_path, 'w', encoding='utf-8') as f:
xml_data = ET.tostring(root_elem, encoding='unicode',
pretty_print=True)
f.write(xml_data)