Temporal compression (also called Temporal downsampling) is a term in Cinematography, a type of model architecture. Compressing video across time (8x stride) so one latent token packs several frames; the reason on-screen typography is destroyed.
Numonic normalises cinematography vocabulary like this so shots, camera moves, and lighting stay searchable and reproducible across AI-generated footage.
Related Terms
See AI Asset Management in Action
Numonic automatically captures provenance, preserves metadata, and makes every AI-generated asset searchable and reproducible.