Deep Diving

Deep Diving is an exploration of the hidden rhythm of knowledge transmission: how machines perceive, interpret, and relate to human presence. The piece was showcased at the Bibliothèque nationale de France during the AI Action Summit in Paris.

Created in collaboration with Mehdi Mejri and Meta, the work combines volumetric captures, AI-driven segmentation, and real footage to create a fragmented digital point of view: a sense of how a machine might see reality.

Using Meta's Segment Anything 2 (SAM2), people in video were detected, isolated, and dynamically integrated into a shifting volumetric environment. AI-generated masks continuously reshaped the 3D space, while custom shaders simulated the unstable perception of a robotic TAD system, as if the machine were actively trying to understand its connection to human presence.

Rather than presenting a single static volumetric state, the scene evolves, fluctuates, and recalibrates, mirroring an ongoing process of comprehension.

The work was introduced alongside Joelle Pineau, VP of AI Research at Meta, and shared with Rachida Dati, French Minister of Culture.

Inspired by the BnF's history, collections, data, and systems for transmitting knowledge. Public coverage highlighted the TAD automated document transport system as a key conceptual reference: a physical network moving books through the library's architecture, reimagined here through machine perception and AI-assisted image processing.