---
title: "Roxana Bujack"
type: "entity"
entity_kind: "person"
status: "hub"
canonical_name: "Roxana Bujack"
aliases: []
first_seen: "2026-07-12T00:00:00.000Z"
writer_model: "claude-sonnet-5"
connects_to: ["non-Riemannian color space","CIELAB","Gaussian curvature (color space)","Deane B. Judd","gradient geometry (cross-domain analogy)"]
---


Los Alamos National Laboratory researcher leading a color-geometry research line distinct from the DFKI Ahrens/Schotten group. Her PNAS 2022 paper (with Teti, Miller, Caffrey & Turton) showed that perceptual color space is not even Riemannian — a Riemannian metric overestimates the perception of large color differences, because perception exhibits diminishing returns that a Riemannian (locally-integrated) metric cannot represent. Her 2025 *Computer Graphics Forum* paper (with Stark, Turton, Miller & Rogers), read directly at capture time, extends this into a non-Riemannian geometric framework and cites [[entity-deane-b-judd|Judd (1968)]] and Silberstein (1938) as the earliest precedents for treating color space as curved — the citation that, read closely, supplied the vault's clearest historical anchor for that lineage.

## References
- [[claim-perceptual-color-space-not-riemannian-bujack-2022]]
- [[claim-judd-1968-silberstein-1938-precede-color-space-curvature-claim]]
- [[observation-cielab-nonuniformity-bridges-gradient-geometry-not-sampling-artifact]]