I successfully ran face detection on my Raspberry Pi 5 using TAPPAS 3.31.0 today.
Building on that, when I tried to use face_attr_resnet_v1_18
to extract attributes like gender and age, I received the following error:
terminate called after throwing an instance of 'std::invalid_argument' what(): No tensor with name face_attr_resnet_v1_18_rgbx/fc3
I used the TAPPAS post-processing library libface_attributes_post.so
. Is this incorrect?