Terminal Blink
The phantasm of an eroding robotic entity before the outage, "the last stream" of its machine visions.
Strokes and paroxysms are often known for abnormal psychedelic visuals, accompanying the attack. Machines may equally be prone to such phantasmagoria, drifting to Computer Envision right before their outage.
Technique / Experiment:
StyleGAN2 neural network, trained on human faces (as the most presented subject in Computer Vision practice), was tuned to Bauhaus architecture, reducing generated identities to impersonal gray-concrete looks. Those 'faceless' faces were processed with another network (StarGAN2), trained on various visual art — from Kandinsky to engravings — to apply diverse imaginative representations. The results have traversed quite far from all the sources, opening the whole new picture.
Conclusion:
Machine Vision is treated hereby as insight (rather than eyesight) concept. Using artificial neural networks as a model of human sensory perception allows to rethink (and possibly redefine) its semantics and aesthetics. Produced synthetic imagery is following familiar artistic tendencies yet; it may go way beyond, as soon as we're ready to welcome that. The original song in soundtrack perceives human life as a single blink between past and future; the human-biased aesthetics may fit even shorter timeframe, as seen by the machine.
-
Vadim EpsteinDirector
-
МифSound
-
Project Type:Animation, Experimental, Music Video, Web / New Media
-
Runtime:2 minutes 37 seconds
-
Completion Date:September 23, 2020
-
Film Color:Color
-
First-time Filmmaker:No
-
Student Project:No
-
NeurIPS Workshop on Machine Learning for Creativity and Designonline
December 12, 2020
Highlights of the year -
18 Festival Transterritorial de Cine UndergroundBuenos Aires
Argentina
November 10, 2022
official selection -
NFT | New Media | Experimental | Digital Arts Film FestivalLondon
United Kingdom
May 28, 2023
Gold Medal in Music Video