[Dataset manager] Fix import for MOTS format #3612
Conversation
I think yes, because this is the only format for which we use |
sizov-kirill
changed the title
sizov-kirill
changed the title
Yes is it, but current CVAT tests check if CVAT can upload annotations dumped by CVAT. |
| from datumaro.components.dataset import Dataset, DatasetItem | ||
| from datumaro.components.extractor import Mask | ||
| from numpy import array |
There was a problem hiding this comment.
What is the point of such import?
There was a problem hiding this comment.
These classes are only needed in this test, so I decided to import them inside the function.
There was a problem hiding this comment.
It's not a significant reason for having them here.
Also, numpy is typically imported as import numpy as np
sizov-kirill
changed the title
| @@ -908,7 +908,7 @@ def _make_image(i, **kwargs): | |||
| attributes["labels"].append({"label_id": idx, "name": label["name"], "color": label["color"]}) | |||
| attributes["track_id"] = -1 | |||
|
|
|||
| dm_item = datumaro.DatasetItem(id=osp.split(frame_data.name)[-1].split('.')[0], | |||
| dm_item = datumaro.DatasetItem(id=osp.splitext(osp.split(frame_data.name)[-1])[0], | |||
There was a problem hiding this comment.
Probably, this should not be in this PR.
There was a problem hiding this comment.
Okay, I'll made these changes in new PR
Motivation and context
Fixed #3360.
Some masks in the MOTS PNG format may contain several polygons with the same track_id, for such annotations to be loaded into CVAT, they must be grouped during export. This is what I did in this PR.
How has this been tested?
Checklist
developbranch[ ] I have increased versions of npm packages if it is necessary (cvat-canvas,cvat-core, cvat-data and cvat-ui)
License
Feel free to contact the maintainers if that's a concern.