video installation, color, 1:1, stereo, be, 2022, 7’14”
a stylegan2 deep learning model is trained on a dataset of different visions of the sublime throughout the ages: the romantic paintings of artists like Caspar David Friedrich, Frederic Edwin Church and William Turner, the films of Andrei Tarkovsky, and contemporary views on the sublime, related to climate breakdown (wildfires, droughts, floods) and pollution. the images reveal a shifted relationship of humanity towards the natural environment. while romantic visions feature an awe-inspiring natural world, contemporary views show the opposite: nature in decline, triggered by human-induced climate disruption and biodiversity loss. the contemporary sublime confronts humanity with its own extinction: beauty and terror unified in a continuous latent interpolation.
dataset creation, model training, latent space navigation, postproduction: Jeroen Cluckers
keys, electronics: Michaël Verlinden
cello, electronics: Jasmijn Lootens
scenography: A. De Causmaecker
programming, patching, video mapping, technical support: Maarten Craeynest
developed during A Matter of Sound, a 2-week residency at the Winter Garden, Royal Academy of Fine Arts, Antwerp (BE), a 2-year residency at ChampdAction, Antwerp (BE), and within the research projects AI-Driven Design (Graphic and Digital Media, department of Science and Technics) and Form a Line into a Circle (MAXlab, Royal Academy of Fine Arts), at AP University of Applied Sciences and Arts Antwerp (BE). with kind support of LUCA School of Arts / KU Leuven, Leuven (BE), docARTES, Ghent (BE) and Immersive Lab, AP University of Applied Sciences and Arts Antwerp (BE).
A Matter of Sound, Winter Garden, Royal Academy of Fine Arts, Antwerp (BE)