Open smedegaard opened 2 years ago
When detecting small objects in large pictures, the label sometimes cover the whole bounding box. Moving the label up gives the same visual semantics and allows to see the object in the box
When detecting small objects in large pictures, the label sometimes cover the whole bounding box. Moving the label up gives the same visual semantics and allows to see the object in the box