This repository has been archived by the owner on Jun 15, 2021. It is now read-only.
Improve hand detection aggregation #22
Labels
enhancement
New feature or request
good first issue
Good for newcomers
help wanted
Extra attention is needed
Currently hand detections are not aggregated, instead the detection with the largest bounding box size is selected. This often leads to incorrect bounding box rotations, which impedes landmark detection.
Accuracy can be approved by implementing non-max suppression or a variation of it.
The text was updated successfully, but these errors were encountered: