Open sfmig opened 1 week ago
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 99.71%. Comparing base (
2e406e1
) to head (299b626
).
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
Issues
2 New issues
0 Accepted issues
Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code
Description
What is this PR
Why is this PR needed? Two small edits are needed in
ValidBboxesDataset
to continue with the bboxes work:Adding
frame_array
toValidBboxesDataset
. Usually running inference on a video using a detection model will output a file (often a csv) that links each detection (aka bounding box) to a frame in the video. As a result, specific frame numbers appear in the output file (see here for a few examples). We would like to be able to pass these specific frame numbers to the bounding boxes dataset if they are defined.Setting default track IDs to 1-based integers. This follows from the discussion at #217.
What does this PR do?
Adds
frame_array
as an optional attribute to theValidBboxesDataset
.frame_numbers
value is a column vector (a numpy array of size (n,1) with n being the number of frames).Sets the default IDs for the
individual_names
to start with 0 ('id_0', 'id_1',...).This PR also includes:
poses
--->ds
.test_sample_data
module to adapt to the new metadata.yaml file (which includes bbox data) that is now in the GIN repo. One fix is temporary (noted with a comment) to ensure the tests pass in this PR.Question: We do not check that the frames are monotonically increasing (i.e. sorted) or that there are no gaps - should we? I think we shouldn't, but let me know thoughts.
References
\
How has this PR been tested?
Tests pass locally and in CI.
Is this a breaking change?
No.
Does this PR require an update to the documentation?
No.
Checklist: