search

Peterisfar / YOLOV3

yolov3 by pytorch
MIT License
196 stars 53 forks source link

Detect too many extra bounding boxes with the test threshold of 0.01. #29

Open wangren97 opened 4 years ago

wangren97 commented 4 years ago

Dear author, When evaluating the mAP of VOC dataset, the performance is much better than darknet YOLOv3 version. However, there are much more extra predicted bounding boxes. The test threshold is default as 0.005 in darknet version and the number of false positives is about 10000~11000. However, in this Pytorch version code, even with a threshold of 0.1, the number of FP is still high. Could you please tell me the reason and how to solve it? Thanks

Peterisfar commented 4 years ago

Dear author, When evaluating the mAP of VOC dataset, the performance is much better than darknet YOLOv3 version. However, there are much more extra predicted bounding boxes. The test threshold is default as 0.005 in darknet version and the number of false positives is about 10000~11000. However, in this Pytorch version code, even with a threshold of 0.1, the number of FP is still high. Could you please tell me the reason and how to solve it? Thanks

how many extra bounding boxes? and how did you calculate？ same images?

wangren97 commented 4 years ago

I tested them both on PASCAL VOC 2007 test set. The extra bounding boxes are counted by total number of boxes minus GT boxes. (consider no false negative, and the total number of detected boxes can be obtained by checking comp4_det_test_category.txt. When checking the txt, you can directly find there are two many extra bounding box.