Getting the space back


In my two previous posts, I experimented with neural-style by taking the fully connected layers into use. This resulted in something quite different, which I have provisionally called neural-mirage. Neural-mirage looks at the uppermost fc layer, the most abstract classification of what the network thinks it sees in the image, and …

Continue reading

I have seen a neural mirage


In my previous post, Controlling image content with FC layers, I described how I had modified neural-style to optimize content based on the classifications obtained from the top-most, fully connected layers (which neural-style does not use).  I had found that the modified program produced visually quite interesting results, but I …

Continue reading