Exploring a Digital Art Collection through Drawing Interactions with a Deep Generative Model

Abstract

Fig. 1. Inferring a Munch-like sketch from a hand-drawn line. From left to right: 1) infrared webcam image from beneath the drawing surface. 2) The cleaned input image. 3) The synthesized sketch. 4) A composite image of the input image over the synthesized sketch. New Snow is an interactive drawing table that investigates human interaction with a deep generative model based on Edvard Munch's sketching practice. Through drawings with pen and paper, the user can interact with the model which will return synthetic sketches based on the input drawings in real-time. The model is a reflection of the training data, and it is thus constrained to representing images within the latent space of Edvard Munch's sketching practice. As the user familiarizes themselves with the model it allows them to become sensitized to the visual aesthetic belonging to this practice. This potential for familiarization with the aesthetic of a dataset via the model has implications for human-AI interaction and non-verbal art mediation. CCS Concepts: • Human-centered computing → Interface design prototyping; Interaction design; Interaction devices; • Applied computing → Fine arts; • Computing methodologies → Machine learning.