search

allanzelener / YAD2K

YAD2K: Yet Another Darknet 2 Keras
Other
2.71k stars 877 forks source link

problem about NMS algroithm #96

Open xiaohu2015 opened 6 years ago

xiaohu2015 commented 6 years ago

I think we shpuld do NMS on the boxes for each class. But using tf.image.non_max_suppression means you treat all classes equally. It may be bad.

tgandor commented 5 years ago

I think it's a trade-off. Plus side: different objects overlapping or 'entangled' can be let through. Minus side: false positives for weaker classes on the same object (like in YOLO3 article addendum, in the discussion about COCO AP metric).