CINEML: Paris

3 minutes | 2021 | Experimental | No subtitles

CineML: Paris employs machine learning to investigate the relationship between filming locations and real-life landmarks, and how cinematic tourism informs the way that people experience and remember the places they visit. I trained several generative adversarial network (GAN) models to analyze public Instagram images from the filming locations of Richard Linklater’s Before Sunset (2004). The GAN models output a series of computer-generated moving images based on the collective memories of Paris. The reconstructed landscape offers a nuanced and data-driven perspective that immerses viewers in a captivating exploration of space and memory.

Lilan Yang (b. 1997, Chongqing, China), is an artist and experimental filmmaker, whose work delves into the flux of migration, the decay of memories and the intricacies of perception. Their work, rooted in 16mm filmmaking, is further enriched by machine learning and data visualization, exploring the materiality and temporality of film. Yang currently lives and works in Boston.