User Tools

Site Tools



This shows you the differences between two versions of the page.

Link to this comparison view

a_neural_algorithm_of_artistic_style [2015/12/17 21:59] (current)
Line 1: Line 1:
 +{{tag>"​Talk Summaries"​ "​Neural Networks"​ "NIPS 2015"​}}
 +====== A Neural Algorithm of Artistic Style ======
 +| Presenter | Leon Gatys |
 +| Context | NIPS 2015 Deep Learning Symposium |
 +| Date | 12/10/15 |
 +This work uses a pre-trained convolutional network (the convolutional part of the VGG network) for image processing (no learning involved). ​ When an image is presented to a convolutional network, it produces a sequence of filtered versions of the input, with irrelevant information discarded although the content is ideally preserved. ​ If the correlation between feature maps in different layers is preserved, the overall texture of the inputs are possible. ​ Optimizing the input (from noise) such that the correlation is the same, texture synthesis is possible. ​ To do style transfer, texture information from a painting is estimated and is combined with a representation of an input image far into the network: ​ First, the correlations are computed for all layers for a painting, then the intermediate representations/​activations for an input image are computed, then the activation correlations for a white noise input are compared to the correlations for the painting, then the highest-layer activation for the white noise is compared (via a Frobenius loss) to those of the query image, and finally backpropagation is used to optimize the pixels so that the correlations and activation loss are jointly minimized. ​ The result is an image which matches the low-level features and colormap of the painting, and gradually gets the content right of the query image. ​ This is, of course, not constrained to artistic style transfer - it can also make two images share some high-level characteristics.
a_neural_algorithm_of_artistic_style.txt ยท Last modified: 2015/12/17 21:59 (external edit)