The document discusses the human eye and vision process. It describes the pinhole camera model of the eye and how it creates a perspective projection of the visual scene onto the retina. It notes that the retina contains different cell types like photoreceptors, bipolar cells, and ganglion cells. The optic nerve transmits signals from ganglion cells to the brain. Rod and cone photoreceptors are described, with rods supporting low-light vision and cones supporting high-resolution color vision. The distribution of rods and cones across the retina is discussed, with a high density of cones in the fovea region for sharp central vision.
The human eye: An introduction to its sensing and imaging processes
1. Sensing and the Imaging Process
(II)
Introduction to Computational and Biological Vision
CS 202-1-5261
Computer Science Department, BGU
Ohad Ben-Shahar
5. The human eye
Pinhole camera model: Basic geometry
Optical axis
Image
plane
Virtual
Image
Pinhole
plane
(center of projection)
6. The human eye
Pinhole camera model: Perspective projection
P = ( x, y , z )
Pi = ( xi , yi , f )
f
Pi = (− xi ,− yi ,− f ) O
x
xi yi f xi = f ⋅ z
OPi = λ ⋅ OP ⇒ λ= = = ⇒
x y z y
yi = f ⋅
z
7. The human eye
Pinhole camera model: Perspective projection
Durer (c. 1500)
8. The human eye
Pinhole camera model: Basic geometry
Recovery of structure is ill defined
13. The human eye
The retina
Variety of cells:
Photoreceptors: ~100 million (95% rods,5% cones)
Bipolar cells
Ganglion cells
Amacrine cells
Horizontal cells
Optic nerve:
Made of axons of secondary (ganglion) neurons
~1 million axons
21. The human eye
The fovea (“pit”)
•Extremely high density of cone receptors
•Few rods. None in the foveola.
•Displaced neural layers
•No blood vessels
•1-1 relationship with bipolar and ganglion cells
1.2 mm