(RE)ROOTING embodies the way the memory of a place is passed on and transformed through our imagination and the tools we use to capture moments in time. Beyond the physical boundaries that the architecture of the National Gallery facade, visions of Singapore’s past and present are transformed into undulating pixel-scapes. These transformed images are created by machine-learning frameworks, which learn the visual patterns of chosen datasets: artworks in the permanent collection of the National Gallery, photographs from the archival library, and images of present-day Singapore.
New images are then generated by the machine that had never existed before, but somehow they feel familiar. The results can be uncanny, non-linear, and also beautiful—similar to the way that we remember past events that we have not personally experienced, by retaining information through archived media and filtering these stories through our own experiences and imagination.
To (RE)ROOT is to connect back to yourself, your community, and the world. It is currently in progress for me—an attempt to rebuild a foundation in my home country in which I had not properly lived for 20 years. The memories I have of Singapore seemed similar to the procedures of creating this piece with machine learning: consuming archived photographs, finding patterns in these fragments of the past, and applying them to my own understanding of the place in which I was born.
Built With
- aftereffects
- premierepro
- runway-ml
- touchdesigner
Log in or sign up for Devpost to join the conversation.