You only look once

计算MAP

https://www.zhihu.com/question/53405779

http://tarangshah.com/blog/2018-01-27/what-is-map-understanding-the-statistic-of-choice-for-comparing-object-detection-models/

http://homepages.inf.ed.ac.uk/ckiw/postscript/ijcv_voc09.pdf

http://host.robots.ox.ac.uk/pascal/VOC/voc2012/htmldoc/devkit_doc.html#sec:ap

code

https://github.com/dmlc/gluon-cv/blob/master/gluoncv/utils/metrics/voc_detection.py

 

Batch Normalization

https://arxiv.org/pdf/1502.03167v1.pdf

top1 top5

[...] where the top-5 error rate is the fraction of test images for which the correct label is not among the five labels considered most probable by the mode.

First, you make a prediction using the CNN and obtain the predicted class multinomial distribution (∑pclass=1∑pclass=1).

Now, in the case of top-1 score, you check if the top class (the one having the highest probability) is the same as the target label.

In the case of top-5 score, you check if the target label is one of your top 5 predictions (the 5 ones with the highest probabilities).

In both cases, the top score is computed as the times a predicted label matched the target label, divided by the number of data-points evaluated.

Finally, when 5-CNNs are used, you first average their predictions and follow the same procedure for calculating the top-1 and top-5 scores.

top1-----就是你预测的label取最后概率向量里面最大的那一个作为预测结果，如过你的预测结果中概率最大的那个分类正确，则预测正确。否则预测错误

top5-----就是最后概率向量最大的前五名中，只要出现了正确概率即为预测正确。否则预测错误。

Multi-scale training

https://arxiv.org/pdf/1805.09300.pdf

 

https://stackoverflow.com/questions/50005852/perform-multi-scale-training-yolov2

 

YOLOv2

https://blog.csdn.net/u010167269/article/details/52638771

https://blog.csdn.net/jesse_mx/article/details/53925356