Intro to Human Visual System Why Should We Be Interested In and Displays Visualization � Fundamental Optics � Hi bandwidth to the brain (70% of all receptors ,40+% of cortex, 4 billion � Fovea neurons) � Perception � Can see much more than we can mentally image � Can perceive patterns (what dimensionality?) These slides were developed by Colin Ware, Univ. of New Hampshire Basic Pathways VENTRAL PATHWAYS Perceptual versus Cultural DORSAL PATHWAYS Faces, Attention, Short term STS IT Eye Movement Control Memory PP A 7A VIP LIP MST FST TEO DP B C PO V3A MT V4 Color and form with color, D Dynamic form attention VISUALL Y GUIDED OBJECT PERCEPTION V3 MOTION PERCEPTION COLOR CONST ANCY TRANSIENT ATTENTION FACES MEMORY V2 SUST AINED Filtering for orientation, color, stereo depth V1
The machinery Human Visual Field 100 A B 80 C 60 Parallel processing D 40 of orientation, texture, Object Identification, 20 color and motion W orking Memory features Detection of 2D patterns, contours and regions LEFT RIGHT Visual Angle Acuities Vernier super acuity (10 sec) θ h Grating acuity r d Two Point acuity (0.5 min)
Human Spatial Acuity Cutoff at 50 cycles/deg. � Receptors: 20 sec of arc � Pooled over larger and larger areas � 100 million receptors � 1 million fibers to brain � A screen may have 30 pixels/cm – need about 4 times as much. � VR displays have 5 pixels/cm Brain Pixels Acuity Distribution 100 80 60 40 20 50 30 10 10 30 50 Distance from Fovea (deg.)
Brain pixels=retinal ganglion cell Pixels and receptive fields 0.8 BP Brain Pixels Field size = 0.006(e+1.0) - Anderson Ganglion cells Characters = 0.046e - Anstis 100 100 1 bp 80 80 60 60 40 40 20 20 50 30 10 10 30 50 50 30 10 10 30 50 Distance from Fovea (deg.) Distance from Fovea (deg.) Small Screen Big Screen 0.2 BP Tartufieri Perception � Many, many ways to trick the vision system.
Intro to Color for Information Display � Color Theory � Color Geometries � Color applications � Labeling � Pseudo-color sequences Trichromacy Cone sensitivity functions Three cones types in retina 100 80 a b 60 40 G+B +R 20 400 500 600 700 W avelength (nm)
Color measurement Short wavelength sensitive cones � Based on the “standard observer” Blue text on a dark background Blue text on a dark background is to be avoided. We have very few is to be avoided. We have very few � CIE tristimulus values XYX short-wavelength sensitive cones in short-wavelength sensitive cones in the retina and they are not very sensitive the retina and they are not very sensitive. � Y is luminance. Chromatic aberration in the eye is also a problem � Assumes all humans are the same Blue text on dark background Blue text on a dark background is to be avoided. We have very few is to be avoided. We have very few short-wavelength sensitive cones in short-wavelength sensitive cones in the retina and they are not very sensitive the retina and they are not very sensitive Color Channels Luminance “channel” Long (red) � Visual system extracts surface information � Discounts illumination level Luminance � Discounts color of illumination Med (green) R-G � Mechanisms Y -B � 1) Adaptation Short (blue) � 2) Simultaneous contrast
Luminance is not Brightness Luminance contrast � Eye sensitive over 9 orders or magnitude � 5 orders of magnitude (room – sunlight) � Receptors bleach and become less sensitive with more light � Takes up to half an hour to recover sensitivity � We are not light meters Contrast for constancy Contrast for constancy
Brightness Lightness and Luminance for Shape-from- Luminance shading � Brightness refers to � Lightness refers to perception of lights perception of surfaces � Brightness non linear � Perceived lightness � Monitor Gamma depends on a reference white Chromatic Channels have Channel Properties Low Spatial Resolution Chromatic Channels � Luminance contrast Luminance Channel Some Natural philosophers T ext on suppose that these colors arise an from accidental vapours diffused needed to see detail � Surfaces of things in the air, which communicate isoluminant � Detail their own hues to the shadows; background so that the colours of the � Labels shadows are occasioned by is hard � Form the reflection of any given sky to read colour: the above observations � Berlin and Kay favour this opinion. � Shading � Categories (about 6- � Motion 10) � Stereo � Red, green, yellow Some Natural philosophers suppose that these colors arise 3:1 recommended from accidental vapours diffused and blue are special in the air, which communicate their own hues to the shadows; 10:1 idea for small text (unique hues)
Color phenomena Color “blindness” � A 3D to a 2D space Chromatic contrast Small field tritanopia � 8 % of males � R-G color blindness a b � Can generate color c blind acceptable d palette � Yellow blue variation OK Implications Color great for classification � Color perception is relative � Rapid visual segmentation � We are sensitive to small differences- � Color helps us hence need sixteen million colors determine type � Not sensitive to absolute values- hence � Only about six we can only use < 10 colors for coding categories green yellow pink white purple red blue brown black orange grey yellow green
Applications Color Coding Large areas: low saturation � Color interfaces � Color coding Small areas high saturation � Color sequences Break isoluminance with borders � Color for multi-dimensional discrete data Color Coding Visual Principles The same rules apply to color � Sensory vs. Arbitrary Symbols coding text and other similar � Pre-attentive Properties information. Small areas should have high saturation colors, � Gestalt Properties � Relative Expressiveness of Visual Cues Large areas should be coded with low saturation colors Luminance contrast should be maintained
Sensory vs. Arbitrary Symbols American Sign Language � Primarily arbitrary, but partly � Sensory: representational � Signs sometimes based � Understanding without training partly on similarity � Resistance to instructional bias � But you couldn’t guess most of � Sensory immediacy them � They differ radically across � Hard-wired and fast languages � Cross-cultural Validity � Sublanguages in ASL are � Arbitrary more representative � Diectic terms � Hard to learn � Describing the layout of a � Easy to forget room, there is a way to indicate by pointing on a plane where � Embedded in culture and applications different items sit. Pre-attentive Processing Example: Color Selection � A limited set of visual properties are processed pre-attentively � (without need for focusing attention). � This is important for design of visualizations � What can be perceived immediately? Viewer can rapidly and accurately determine � What properties are good discriminators? whether the target (red circle) is present or absent. Difference detected in color. � What can mislead viewers? All Preattentive Processing figures from Healey 97 http: / / www.csc.ncsu.edu/ faculty/ healey/ PP/ PP.html
Pre-attentive Processing Example: Shape Selection � < 200 - 250ms qualifies as pre-attentive � eye movements take at least 200ms � yet certain processing can be done very quickly, implying low-level processing in parallel � If a decision takes a fixed amount of time regardless of the number of distracters, it Viewer can rapidly and accurately determine is considered to be pre-attentive. whether the target (red circle) is present or absent. Difference detected in form (curvature) Example: Conjunction of Features Example: Emergent Features Viewer cannot rapidly and accurately determine whether the target (red circle) is present or absent when target has two or more features, each of which are Target has a unique feature with respect to present in the distractors. Viewer must search sequentially. distractors (open sides) and so the group All Preattentive Processing figures from Healey 97 can be detected preattentively. http://www.csc.ncsu.edu/faculty/healey/PP/PP.html
Asymmetric and Graded Preattentive Properties Example: Emergent Features � Some properties are asymmetric � a sloped line among vertical lines is preattentive � a vertical line among sloped ones is not � Some properties have a gradation � some more easily discriminated among than others Target does not have a unique feature with respect to distractors and so the group cannot be detected preattentively. Use Grouping of Well-Chosen Shapes for Displaying Multivariate Data SUBJECT PUNCHED QUICKLY OXIDIZED TCEJBUS DEHCNUP YLKCIUQ DEZIDIXO CERTAIN QUICKLY PUNCHED METHODS NIATREC YLKCIUQ DEHCNUP SDOHTEM SCIENCE ENGLISH RECORDS COLUMNS ECNEICS HSILGNE SDROCER SNMULOC GOVERNS PRECISE EXAMPLE MERCURY SNREVOG ESICERP ELPMAXE YRUCREM CERTAIN QUICKLY PUNCHED METHODS NIATREC YLKCIUQ DEHCNUP SDOHTEM GOVERNS PRECISE EXAMPLE MERCURY SNREVOG ESICERP ELPMAXE YRUCREM SCIENCE ENGLISH RECORDS COLUMNS ECNEICS HSILGNE SDROCER SNMULOC SUBJECT PUNCHED QUICKLY OXIDIZED TCEJBUS DEHCNUP YLKCIUQ DEZIDIXO CERTAIN QUICKLY PUNCHED METHODS NIATREC YLKCIUQ DEHCNUP SDOHTEM SCIENCE ENGLISH RECORDS COLUMNS ECNEICS HSILGNE SDROCER SNMULOC
Recommend
More recommend
