You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Brian Wandell edited this page Aug 22, 2026
·
1 revision
Human Eye (sceneEye)
sceneEye renders through a schematic human-eye optical model instead of a
camera lens — Navarro, Arizona, and LeGrand models are built in. It always
renders on the CPU, regardless of your Docker GPU configuration; see
docs/sceneEye-gpu.md
for why. For the full API — accommodation, retinal geometry, lens density,
the pinhole debug mode — see the iset3d-sceneEye skill in the ISET3D
repository.
Tutorials
There is currently no runnable, introductory sceneEye tutorial in the
repository. tutorials/sceneEye/ has one advanced analysis script, and
examples/sceneEye/ contains only underDevelopment/ scripts, which the
repository's own convention treats as exploratory rather than first-copy
templates:
t_retinaSamplingTest.m
— retinal sampling; an analysis script, not an introductory walkthrough.
Until an introductory tutorial exists, start from the iset3d-sceneEye
skill's code examples directly.