a2c [1910.11118] Shallow Art: Art Extension Through Simple Machine Learning

Enable JavaScript to see more content

arXiv2code // top new 14d 1m 2m 3m // Enable JavaScript to see more content

[share]	Go to arXiv Download as Jupyter Notebook	2019-10-25
[1910.11118] Shallow Art: Art Extension Through Simple Machine Learning Through the implementation of a wrapper method which can utilize any single-output classification or regression model to learn from and complete images, Shallow Art presents a novel new method of computational art generation which straddles the boundary between computational creativity and creativity support Abstract: Shallow Art presents, implements, and tests the use of simple single-output classification and regression models for the purpose of art generation. Various machine learning algorithms are trained on collections of computer generated images, artworks from Vincent van Gogh, and artworks from Rembrandt van Rijn. These models are then provided half of an image and asked to complete the missing side. The resulting images are displayed, and we explore implications for computational creativity. ‹Figure 2: Each pixel on the right is predicted using the values of all pixels on the left (attributes). This process is repeated so that each pixel on the right half of the image is the output of one independent model. (Van Gogh & Rembrandt)Figure 3: Horizontal Lines. Images created by classification Wrapper-Models trained on 50 randomly generated training images as described in ??. (Circles)Figure 4: Vertical Lines. Images created by classification Wrapper-Models trained on 50 randomly generated training images as described in ??. (Circles)Figure 5: Circles. Images created by classification WrapperModels trained on 50 randomly generated training images as described in ??. (Circles)Figure 6: Triangle. Images created by classification WrapperModels trained on 50 randomly generated training images as described in ??. (Circles)Figure 7: Triangle WM. Outputs from a decision tree WM trained on 1900 images of triangles as described in ?? (Circles)Figure 8: Vincent van Gogh WM. Outputs from a decision tree WM trained on 1900 van Gogh works as described in ?? (Circles)Figure 9: Rembrandt van Rijn WM. Outputs from a decision tree WM trained on 900 Rembrandt works as described in ?? (Circles)›