tensorrt-ssd-easy issues

cmake successfully,but make return an error

63

Thanks for your sharing. First,I download your code,and build a folder named "build". Second, I enter the build folder and cmake ../ it. It works well. Third, I make it,but...

JingliangGao

Recognition effect is different after using tensorRT

What should I do when using tensorRT with a different recognition effect than when using caffe_ssd?

Yangxiu123321

environment issue and createSSDDetectionOutputPlugin issue

1

hi saikumarGadde Can you describe the software environment? i want to use it on px2 , thank for u help .

allenchen8210

Mobilenet implementation

14

Hello, i tried to deploy mobilenet-ssd (https://github.com/chuanqi305/MobileNet-SSD), and have a problem with the output shape "OUTPUT1 Tensor Shape is: C: 21 H: 1917 W: 1" It should be 1917 21...

NguyenHongHanh

the result is sometimes different infer the same image

1

the result is close to result on caffe, but the result is sometimes different infer the same image. for example: cf box 0.997599 , 0.431316 , 0.270604 , 0.792519 ,...

huang8705022

error

5

I run this code,the got an error /tensorNet.cpp:39: void TensorNet::caffeToTRTModel(const string&, const string&, const std::vector&, unsigned int): Assertion `blobNameToTensor != nullptr' failed.

lucheng07082221

Do I need to implement softmax layer myself or I can run this code directly?

4

In Chen1992's code,I need to implement softmax myself and in this code do I need to do so as well or I can run it directly? Any replies are apprciated,thx!!!

zbw4034

Why the confidence of the prediction will exceed 1?

2

I changed the number of categories(30) and the proportion of boxes(add 3 aspect ratio in the 4_3 layer,and 4 in the fc7 and conv6_2 layer). I changed all the 21...

408550969