Skip to content

Commit

Permalink
hierarchical labeling
Browse files Browse the repository at this point in the history
  • Loading branch information
@bonhunko
bonhunko committed Oct 18, 2022
1 parent 5ff089d commit 4ca506e
Show file tree
Hide file tree
Showing 12 changed files with 413 additions and 3 deletions.
2 changes: 2 additions & 0 deletions CHANGELOG.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Added
- Add jupyter sample introducing how to merge datasets
(<https://github.com/openvinotoolkit/datumaro/pull/738>)
- Support for hierarchical labeling with single-selection
(<https://github.com/openvinotoolkit/datumaro/pull/742>)

## 06/09/2022 - Release v0.3.1
### Added
Expand Down
9 changes: 7 additions & 2 deletions datumaro/components/annotation.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -97,6 +97,7 @@ class Category:
name: str = field(converter=str, validator=not_empty)
parent: str = field(default="", validator=default_if_none(str))
attributes: Set[str] = field(factory=set, validator=default_if_none(set))
single_selection: bool = field(default=False, validator=default_if_none(bool))

items: List[str] = field(factory=list, validator=default_if_none(list))
_indices: Dict[str, int] = field(factory=dict, init=False, eq=False)
Expand Down Expand Up @@ -146,13 +147,17 @@ def _reindex(self):
self._indices = indices

def add(
self, name: str, parent: Optional[str] = None, attributes: Optional[Set[str]] = None
self,
name: str,
parent: Optional[str] = None,
attributes: Optional[Set[str]] = None,
single_selection: Optional[bool] = False,
) -> int:
assert name
assert name not in self._indices, name

index = len(self.items)
self.items.append(self.Category(name, parent, attributes))
self.items.append(self.Category(name, parent, attributes, single_selection))
self._indices[name] = index
return index

Expand Down
37 changes: 37 additions & 0 deletions datumaro/plugins/datumaro_format/converter.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@

from datumaro.components.annotation import (
Annotation,
AnnotationType,
Bbox,
Caption,
Cuboid3d,
Expand Down Expand Up @@ -279,6 +280,7 @@ def _convert_label_categories(self, obj):
"name": cast(label.name, str),
"parent": cast(label.parent, str),
"attributes": self._convert_attribute_categories(label.attributes),
"single_selection": cast(label.single_selection, bool),
}
)
return converted
Expand Down Expand Up @@ -336,6 +338,7 @@ def apply(self):

for item in self._extractor:
subset = item.subset or DEFAULT_SUBSET_NAME
item = self._filterout_for_single_selection(item, self._extractor.categories())
writers[subset].add_item(item)

for subset, writer in writers.items():
Expand Down Expand Up @@ -383,3 +386,37 @@ def patch(cls, dataset, patch, save_dir, **kwargs):
)
if osp.isdir(related_images_path):
shutil.rmtree(related_images_path)

@staticmethod
def _filterout_for_single_selection(item, categories):
name2parent_ss = {}
for label_category in categories[AnnotationType.label]:
name2parent_ss[label_category.name] = (
label_category.parent,
label_category.single_selection,
)

# collect childrens that have the same parent which only allow single-selection
parent2ss_indices = {}
for i, annotation in enumerate(item.annotations):
if annotation._type != AnnotationType.label:
continue

label_name = categories[AnnotationType.label][annotation.label].name

parent, single_selection = name2parent_ss[label_name]

if len(parent) == 0:
continue

if not single_selection:
parent2ss_indices[parent] = parent2ss_indices.get(parent, []) + [i]

# remove labels that dis-obey the single-selection rule
for indices in parent2ss_indices.values():
if len(indices) <= 1:
continue
for i in sorted(indices, reverse=True):
del item.annotations[i]

return item
5 changes: 4 additions & 1 deletion datumaro/plugins/datumaro_format/extractor.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -65,7 +65,10 @@ def _load_categories(parsed):
label_categories = LabelCategories(attributes=parsed_label_cat.get("attributes", []))
for item in parsed_label_cat["labels"]:
label_categories.add(
item["name"], parent=item["parent"], attributes=item.get("attributes", [])
item["name"],
parent=item["parent"],
attributes=item.get("attributes", []),
single_selection=item.get("single_selection", False),
)

categories[AnnotationType.label] = label_categories
Expand Down
81 changes: 81 additions & 0 deletions tests/assets/datumaro_dataset/annotations/test.json
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,81 @@
{
"info": {},
"categories": {
"label": {
"labels": [
{
"name": "manmade",
"parent": "",
"single_selection": true,
"attributes": []
},
{
"name": "car",
"parent": "manmade",
"single_selection": false,
"attributes": []
},
{
"name": "bicycle",
"parent": "manmade",
"single_selection": false,
"attributes": []
},
{
"name": "accesory",
"parent": "",
"single_selection": false,
"attributes": []
},
{
"name": "glasses",
"parent": "accesory",
"single_selection": false,
"attributes": []
},
{
"name": "glove",
"parent": "accesory",
"single_selection": false,
"attributes": []
}
],
"attributes": []
}
},
"items": [
{
"id": "c",
"annotations": [
{
"id": 0,
"type": "label",
"attributes": {
"score": 1.0
},
"group": 0,
"label_id": 1
},
{
"id": 0,
"type": "label",
"attributes": {
"score": 1.0
},
"group": 0,
"label_id": 3
}
],
"image": {
"path": "../tests/assets/datumaro_dataset/images/test/c.jpg",
"size": [
10,
5
]
},
"media": {
"path": "../tests/assets/datumaro_dataset/images/test/c.jpg"
}
}
]
}
107 changes: 107 additions & 0 deletions tests/assets/datumaro_dataset/annotations/train.json
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,107 @@
{
"info": {},
"categories": {
"label": {
"labels": [
{
"name": "manmade",
"parent": "",
"single_selection": true,
"attributes": []
},
{
"name": "car",
"parent": "manmade",
"single_selection": false,
"attributes": []
},
{
"name": "bicycle",
"parent": "manmade",
"single_selection": false,
"attributes": []
},
{
"name": "accesory",
"parent": "",
"single_selection": false,
"attributes": []
},
{
"name": "glasses",
"parent": "accesory",
"single_selection": false,
"attributes": []
},
{
"name": "glove",
"parent": "accesory",
"single_selection": false,
"attributes": []
}
],
"attributes": []
}
},
"items": [
{
"id": "a",
"annotations": [
{
"id": 0,
"type": "label",
"attributes": {
"score": 1.0
},
"group": 0,
"label_id": 0
}
],
"image": {
"path": "../tests/assets/datumaro_dataset/images/train/a.jpg"
},
"media": {
"path": "../tests/assets/datumaro_dataset/images/train/a.jpg"
}
},
{
"id": "b",
"annotations": [
{
"id": 0,
"type": "label",
"group": 0,
"label_id": 0
},
{
"id": 1,
"type": "label",
"group": 0,
"label_id": 1
},
{
"id": 2,
"type": "label",
"group": 0,
"label_id": 2
},
{
"id": 3,
"type": "label",
"group": 0,
"label_id": 5
}
],
"image": {
"path": "../tests/assets/datumaro_dataset/images/train/b.jpg",
"size": [
2,
8
]
},
"media": {
"path": "../tests/assets/datumaro_dataset/images/train/b.jpg"
}
}
]
}
58 changes: 58 additions & 0 deletions tests/assets/datumaro_dataset/annotations/validation.json
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,58 @@
{
"info": {},
"categories": {
"label": {
"labels": [
{
"name": "manmade",
"parent": "",
"single_selection": true,
"attributes": []
},
{
"name": "car",
"parent": "manmade",
"single_selection": false,
"attributes": []
},
{
"name": "bicycle",
"parent": "manmade",
"single_selection": false,
"attributes": []
},
{
"name": "accesory",
"parent": "",
"single_selection": false,
"attributes": []
},
{
"name": "glasses",
"parent": "accesory",
"single_selection": false,
"attributes": []
},
{
"name": "glove",
"parent": "accesory",
"single_selection": false,
"attributes": []
}
],
"attributes": []
}
},
"items": [
{
"id": "d",
"annotations": [],
"image": {
"path": "../tests/assets/datumaro_dataset/images/validation/d.png"
},
"media": {
"path": "../tests/assets/datumaro_dataset/images/validation/d.png"
}
}
]
}
Binary file added tests/assets/datumaro_dataset/images/test/c.jpg
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added tests/assets/datumaro_dataset/images/train/a.jpg
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added tests/assets/datumaro_dataset/images/train/b.jpg
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added tests/assets/datumaro_dataset/images/validation/d.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading

0 comments on commit 4ca506e

Please sign in to comment.