Resource retrieval

Get the pre-trained net:

In[1]:=

Out[1]=

Basic usage

This net outputs a 64x64 heat map for each of the 68 landmarks:

In[2]:=

Obtain the dimensions of the heat map:

In[3]:=

Out[3]=

Visualize heat maps 1, 12 and 29:

In[4]:=

Out[4]=

Evaluation function

Write an evaluation function that picks the maximum position of each heat map and returns a list of landmark positions:

In[5]:=

Landmark positions

Get the landmarks using the evaluation function. Coordinates are rescaled to the input image size so that the bottom-left corner is identified by {0, 0} and the top-right corner by {1, 1}:

In[6]:=

Out[6]=

Group landmarks associated with different facial features by colors:

In[7]:=

Visualize the landmarks:

In[8]:=

Out[8]=

Preprocessing

The net must be evaluated on facial crops only. Get an image with multiple faces:

In[9]:=

Write an evaluation function that crops the input image around faces and returns the crops and facial landmarks:

In[10]:=

Evaluate the function on the image:

In[11]:=

Out[11]=

Visualize the landmarks:

In[12]:=

Out[12]=

Robustness to facial crop size

Get an image:

In[13]:=

Crop the image at various sizes:

In[14]:=

Out[14]=

Inspect the network performance across the crops:

In[15]:=

Out[15]=

Net information

Inspect the number of parameters of all arrays in the net:

In[16]:=

Out[16]=

Obtain the total number of parameters:

In[17]:=

Out[17]=

Obtain the layer type counts:

In[18]:=

Out[18]=

Display the summary graphic:

In[19]:=

Out[19]=

Export to MXNet

Export the net into a format that can be opened in MXNet:

In[20]:=

Out[20]=

Export also creates a net.params file containing parameters:

In[21]:=

Out[21]=

Get the size of the parameter file:

In[22]:=

Out[22]=

The size is similar to the byte count of the resource object:

In[23]:=

Out[23]=