how to understand this
Opened this issue · 0 comments
sqdcbbj commented
if all the training samples are equally treated, there will be a misalignment between the two heads: the location with the highest category score is usually not the best position for regressing the object boundary.