Closed bkj closed 6 years ago
Hi,
Thank you for your interest in our work.
The problem you are describing is a know issue caused since the newer Caffe versions are not 100% compatible with the version used in our project.
For a detailed description of the problem and a solution, see the project page (update from the 15th of July): http://www.openu.ac.il/home/hassner/projects/cnn_agegender/
Best, Gil
Ah great thank you.
Also, do you know of any work that actually does face detection in images that I could use alongside your software. At the moment, if I get the network a picture of a street sign, it gives it a 100% male label. Ideally we'd want to detect age and gender of a face IFF a face is present in the image.
Hi,
You need to look for face detection code. There are many works on the subject, for example the following: https://www.ics.uci.edu/~xzhu/face/
Best, Gil
Hi, I also run to the same issue and I was wondering:
which Caffe version did you use for the project?
Thanks!
Hi @IvonaTau ,
I'm afraid I don't remember which version we used. We used the latest source code from Caffe's github repository at that time (around February 2015).
Gil
When I try to run the code from the notebook, I get the following error
which suggests that the
mean
array needs to be reshaped for input intocaffe.Classifier
. Any suggestions on how to proceed?