You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the current implementaion, one has to initialize MeanAveragePrecision() twice to get bbox and segmentation results. The implementation should support getting both bbox and segmentation metrics with a single initialization.
Motivation
The pytorch evaluation has the possibility to get both bbox and segmentation mAP in one run. This implementation has been out for a while now and I believe we should have a similar behaviour!
Pitch
Rather than taking a Literal of either a bbox or segm, the input argument iou_types (instead of iou_type) should take a list instead. The returned value should Dict but with first key being the iou_type.
Please let me know what you think!
The text was updated successfully, but these errors were encountered:
🚀 Feature
In the current implementaion, one has to initialize MeanAveragePrecision() twice to get bbox and segmentation results. The implementation should support getting both bbox and segmentation metrics with a single initialization.
Motivation
The pytorch evaluation has the possibility to get both bbox and segmentation mAP in one run. This implementation has been out for a while now and I believe we should have a similar behaviour!
Pitch
Rather than taking a Literal of either a bbox or segm, the input argument
iou_types
(instead ofiou_type
) should take a list instead. The returned value should Dict but with first key being the iou_type.Please let me know what you think!
The text was updated successfully, but these errors were encountered: