Track 1 Source Atlas
Where literal substrate disagrees with library-free structure-space — and which dimensions hold any two apart.
Every image is a displacement vector from a fixed geometric origin, across the full VCLIG battery, scaled by each axis's geometric range — no learned population. NCD asks whether two images share compressible substrate. VTL asks whether they organize space similarly. The useful cases are often the disagreements.
Parallax Metrology
Track 1 Source Atlas
24images analyzed
276pair crossings
78structural kinship cases
0.8053median VTL distance
The atlas separates literal substrate similarity from structural kinship. The strongest cases are the lower-right disagreement pairs: high NCD, low VTL. Those are images that do not compress together, but still organize space similarly.
In this run, the dominant recurring drivers were highlight mass, shadow mass, void count, vertical centroid, gestural thrust. Use the pair detail panel below to see which dimensions explain any specific separation.
Attribution: All images: MidJourney
Collection summaryhow the full set divides across the NCD / VTL crossing
Distance matrixdarker green = structurally closer · darker red = farther · click a cell for drivers
structural kinship case — far by substrate (NCD), near by structure (VTL)