Error reading facenet model with dnn module

asked 2019-06-14 11:23:30 -0600

Hi,

I'm working on a face recognition project with opencv. I use pretrained model for facenet from this github https://github.com/davidsandberg/facenet (model 20180402-114759) and load with readNetFromTensorflow. All I did is download the model and run those codes:

detector = cv2.dnn.readNetFromTensorflow(20180402-114759.pb)

However, I got error: error: (-2:Unspecified error) Input layer not found: image_batch in function 'connect'

Have anyone had this problem before? What did you do to solve it?

Thanks

edit retag flag offensive close merge delete

add a comment

1 answer

Sort by » oldest newest most voted

answered 2019-12-08 10:30:05 -0600

tanfulun
21 ●1 ●1

hi, I have search around about 'facenet model deploy on cv-dnn', but get nothing useful. So, I try to do it by myself, and manage to run face feature extraction on cv-dnn. I put my work here "https://github.com/TanFluent/facenet_...". It is a little late but may help others. LOL