residual
residual draws from early experimental video art and analog feedback systems while reflecting contemporary concerns around surveillance, algorithmic mediation, and machine perception. Using a live camera feed projected back onto itself, the work creates recursive visual patterns that continuously evolve through the presence and movement of participants. Rather than positioning technology as autonomous or inevitable, the work treats it as collaborative and something shared through interaction, disruption, and participation.
UCSB House Show 2025
My team and I created this installation for UCSB's House Show, an artist showcase featuring work from the Media Arts & Technology, Music, and Art departments. The piece uses computer vision to track people as they move through the space, whether they're passing by or intentionally interacting with it. Their movement becomes a paintbrush, drawing lines across a digital canvas while the program continuously cycles through colors and randomly varies the thickness of each stroke.
My primary contribution was developing the interaction layer between the computer vision system and the visual output. I mapped the tracked movement data to the brushstroke behaviors, translating participants' positions and motion into the different visual elements that shape each composition. I also helped design the data mappings that transform the generated drawings into sound.
Every minute, the generated image is saved and transformed into an audio composition. The system maps the y-coordinate of each point along a brushstroke to a frequency, so higher positions produce higher pitches while lower positions create deeper tones. The duration of each note is determined by the length of its corresponding brushstroke. Once the audio playback is complete, the canvas is cleared and the process begins again.
The result is a new audiovisual artwork every minute, shaped collectively by everyone who passes through the space, whether they intended to participate or not.
Using p5.js and ml5.js, I developed an interactive AR experience where a swarm of bees targets and follows a participant. When there's no participant, the bees move slowly across the screen. Once a participant is detected, the bees swarm and track the participant's movements until they are no longer within the camera’s view.