No, I am not saying that neural networks can achieve everything. The endless possibilities I am facing do not extend outwards towards practical goals. Instead, when I am experimenting with applications of neural network for creative and original art, I am overwhelmed by the endless options and variations at my …
The core of neural-style is the pre-trained neural network. Pre-trained means that the net has been trained using a large number of images, usually with the goal that the net be able to recognize objects and features in images. The default neural network for neural-style, VGG-19 is very effective for …
The big thing about neural-style is being able to separate style and content of an image, and to synthesize new images mixing content and style from different sources. The original paper describing this new technology demonstrated how it is possible to apply the style of famous paintings from Van Gogh, Kandinsky …
In June this year, I visited the Kumu art museum in Tallinn, Estonia, to see a small exhibition about metamorphoses of Malevitch’ Black Square in Estonian art. The exhibition was small and there was nothing particularly exciting about it, apart from the fact the square had clearly both inspired and …
What if a Processing sketch would be monitoring a webcam stream, painting and updating an image with vertical stripes. Something like this. What is going on here? The sketch is drawing vertical stripes every 16 pixels. It is looking at the brightness of the video image at each point, and …
The style image in neural-style does not have to be a work of art. In this example, I will take a part of a topographic map, like this and apply it to two photographs, a building and a street view. (source: https://tiedostopalvelu.maanmittauslaitos.fi/tp/kartta) The first save, after 100 iterations, is essentially …
As we have seen, neural networks can create great pictures, but the process is still too slow to be applied in real-time or to implement responsive applications in which the user can affect the resulting image while it is being generated. The Processing framework and programming language shines in such applications. …
The same photo as before, but now the style is taken from a graphic print by Saara Mesterton and Johanna Töyrylä: The style image is given added weight. Then, after 100 and 200 iterations, respectively, we get these images: Somehow I find these early iterations fascinating, although others seem to …
I returned to the photo from Sarvikallio, Tuusula, Finland, and applied to it the style from an abstract painting by Saara Mesterton. …which resulted in this: But what if the style and content images are swapped? That we copy the style of the photograph into the content of the …
After covering the basics of using neural-style, I ventured into making my own portrait, based on a photo taken when I was defending my PhD thesis. But to start with, the style was taken not from any painting but from some text in Yiddish. The immediate result, after 100 iterations …
