PKUZHOU / MTCNN_FaceDetection_TensorRT

MTCNN C++ implementation with NVIDIA TensorRT Inference accelerator SDK
202 stars 57 forks source link

why add means? #18

Closed xiexuliunian closed 4 years ago

xiexuliunian commented 4 years ago

image in this code,why add mean? not sub?

PKUZHOU commented 4 years ago

Where is the location of this code?

xiexuliunian commented 4 years ago

image

PKUZHOU commented 4 years ago

It seems to be a mistake...

xiexuliunian commented 4 years ago

yeah,but the result seem also not too bad......lol

PKUZHOU commented 4 years ago

Deep learning's magic