2023-2024
'self-contained 009.1' exhibited at Colección SOLO, photos courtesy of Colección SOLO.
All that we consider living around us is partly the result of expressing the organic encoding of information in DNA form. The series ‘self-contained’ takes inspiration from the phenomena of encoding & decoding, a process rife with mutations. Coupled with the evolutionary nature of the crossbreeding and random changes, this mirrors some of the memetic aspects of our digital spaces wherein images are often combined, edited, degraded or re-contextualised.
In this series, target images becomes the subject of an iterative crossbreeding of random samples of strings from other compressed images, leading to a mutative evolution. Akin to grafting two plants together here images from a dataset are randomly spliced into the image “genome”. Furthermore rudimentary a-life simulations are utilised to explore notions of "error correction" in the mutative process, tentatively tracing the contours of the complexities of natural processes. Decoding regenerates the image to a more recognisable form which is then aided by neural networks trained on the same dataset to further return colour and detail.
It is this iterative, simultaneous process of compression-decompression that the works depict. Which, as around us, occurs in shared simultaneity. Just as our cells divide, mutate and grow, so do the contents of our shared digital spaces as we remix, are remixed, hyped, and made obsolete.
With the work 'self-contained 009.x' another degree of self-containment is achieved in the synthesis and storage of a compressed version of the work in DNA form. The DNA is stored in a capsule fitted into a custom sculptural container. Thus the work explores a full-circle expression of digital-physical intertwining from the application of simulated natural processes to the process of image-making to the final re-encoding into physical, synthesised DNA.
Stage one: a target image becomes the subject of a iterative crossbreeding of random samples of strings from other compressed images, leading to a mutative evolution. Akin to grafting two plants together here images from a dataset are randomly spliced into the image "genome".
Stage two: once encoded and then mutated, a parallel process of decoding regenerates the image to a more recognisable form, aided by neural networks trained on the same dataset to further regenerate colour and detail.
Exhibition at Neuehouse LA, photos by Kenny Laubbacher
metadata:
medium: digital video (generative code, neural networks), synthesised DNA in metal capsule, sculptural capsule holder (stainless steel, 3D printed)
dimensions: various
duration: 1 min 30s
frame rate: 4fps
codec: Apple ProRES 422 HQ MOV / h.264 MP4 (archival / compatible)
metadata:
medium: generative code, neural networks
dimensions: 2752 × 4892 (9:16) / 4892 × 2752 (16:9)
frame rate: 8fps
codec: Apple ProRES 422 HQ MOV / HEVC MOV / h.264 MP4 (archival / high quality / compatible)