System information (version)

• OpenCV => 3.4.0
• Operating System / Platform => Windows 10 64 Bit
• Compiler => Visual Studio 2015
• Cmake => 3.9.6

Detailed description

I retrained Mobilenet-SSD with my own data strictly follow this guidance,
Then I test my retrained model with python and caffe in ubuntu ,get the right detection result.

However, when I use the retrained model in OPENCV3.4(windows 10,vs2015), it can't detect any object.

Steps to reproduce

This is my python script used in caffe test.

This is the C++ script used in opencv test.

What do I have to solve this problem?

I am looking forward to your reply.
Thank you!
I have tested with different inScaleFactor 、meanVal 、RGB/grayscale 、 resize method , none of them can detect any object.