FasterRCNN-Encapsulation-Cplusplus
FasterRCNN-Encapsulation-Cplusplus copied to clipboard
MeanBinary params
Thank you for providing such a good instruction of using faster rcnn with C++. I have some question . 1: MAT to blob ,why the mean params is (102.9801) (115.9465),(122.7717)。 2: I use the model provide by the RGB. but the result of the C++ And python had great difference. the boxes is wrong and there are much boxes。 could you tell me how to solve these problems.
I have the same question about the mean parameters. @zhanglaplace were you able to figure out the answer?
@alex-bot In my opinion, the mean parameters need to be set according to our own model.