The development of this piece spurred from my work and research with the TRELS Summer Research Program during the summer of 2021. My thesis at this point involved examining codified narratives in dance performance in an attempt to analyze and quantify movement through computer vision techniques and procedurally generated computer graphics. This period of development involved studying the semiotics of dance and movement, understanding concepts such as Labanotation and Laban Movement Analysis, and using pose estimation to classify movement and record skeletal key point data from a moving figure. The current stage of my work uses pose estimation and keypoint detection to track movement and classify poses in real time, from which a procedurally generated particle system is created and manipulated in Max/MSP according to the performer’s movement and the particular scene of the narrative. The work flow for this project begins with live camera feed, which is translated into keypoint and classification data using a model I trained in Google’s Teachable Machine. The data is then routed into a Max/MSP patch which generates the projected particle system behind the dancer as seen in the recorded performance.

I am incredibly proud of the learning, discovery, and collaboration that involved making this piece; and the opportunity to return to a live performance setting after two years of isolation and remote work meant so much to me. The debut of this project for my senior thesis is the beginning of an ongoing exploration into these techniques, questions, and ideas in my process as an artist, and I am excited to continue to rework and refine these concepts for future iterations. Special thank you to my professors Amy Alexander and Tom Erbe for their help and guidance during this process, Karen Makhoul for starting this journey with me, and Kenzie Barnickle for her talent and choreographic vision in helping me realize this performance.