Stimulus Material

Interdisciplinary perception research

Function of stimulus material

Controlled stimulus material is needed when visual judgements are studied systematically.

The material basis goes back to 3D acquisitions from 2005 onward. It links morphological prior work on faces and age with questions of visual judgement, observer variability, and confidence.

Systematic control or variation of age, view, image information, and task makes it possible to distinguish more specifically between potential contributions of the face, the view, the image conditions, and the observer.

208people
9–85years
3 scansper person · left, frontal, right
624position-specific single scans

How does a visual judgement change when the same person is represented differently?

In faces, view, projection, and image conditions are not secondary. They affect which cues are visible, which are obscured, and what observers can infer from them.

Stimulus pool and age range

The stimulus pool comprises 3D face data and standardised image data from 208 people (105 women, 103 men) aged 9 to 85 years. The data allow controlled views, derived images, age-estimation tasks, comparison tasks, and analyses of visible cues.

The point is not to demonstrate scanning technology. The material provides a basis for relating stimulus variation, observer judgement, and uncertainty in a systematic way.

Schematic age distribution of the 3D stimulus material
Schematic overview of the age range represented in the 3D stimulus material.

What can be studied

Age and visible cues

Which structures are used for age estimates, and why do estimates vary more for some people than for others?

View and projection

How do frontal, lateral, or derived views change the available information and the confidence of judgements?

Observer judgements

How consistent are judgements across observers, across repeated presentations, and in relation to subjective confidence?

Workflow diagram: 3D acquisition, standardised views, age and comparison tasks, judgement and confidence data.
The material links controlled image derivations with age and comparison tasks as well as judgement and confidence data.

3D acquisition as a basis

3D acquisition separates geometric shape, surface texture, and derived 2D views. This makes differences between facial form, image representation, and observer judgement easier to examine directly.

Technical details on acquisition

The data were acquired with a non-contact 3D laser scanner based on laser triangulation. The scanner does not interpret features; it records a dense point cloud of the object surface. These points were subsequently combined with photographic RGB information.

Each scan captured 640 × 480 measurement values, corresponding to 307,200 measurement points. The original technical documentation reports a measurement accuracy of ±0.010 mm at one sigma and a geometric z-resolution of 0.008 mm.

The scan itself took 2.5 seconds per position. For each head, three positions were acquired: lateral left, frontal, and lateral right. Focus distance was 1060 mm for the frontal scan and 920 mm for the lateral scans.

ScannerKonica Minolta VI-910
MethodNon-contact laser triangulation
Values640 × 480 = 307,200 points
Raw material208 × 3 scans = 624 position-specific 3D scans
Accuracy±0.010 mm at 1 sigma
Z-resolution0.008 mm
Scan time2.5 seconds per position
Positionslateral left · frontal · lateral right
TextureRGB, 24-bit color depth
Principle of a triangulation sensor
Principle of laser triangulation used for 3D acquisition.

Documentation of the 3D workflow

Workflow figures

Prior work on visible ageing cues

Labeled overview of visible wrinkle, furrow, and line structures
Schematic overview of visible wrinkle, furrow, and line structures. Drawing: Kathrin Hirthammer, based on specifications by Bastian J. Hirthammer. The figure draws on earlier morphological work and is not a definitive current category system.
Show additional views and legend

The prior work included frontal, lateral, and oblique isometric views. This material makes it possible to examine to what extent judgements of visible ageing cues depend on view, contour, skin relief, and soft-tissue form.

a

Lineae frontales transversae — horizontal forehead lines

b

Lineae glabellares verticales — vertical glabellar lines

c

Sulcus orbitalis superior — superior orbital furrow

d

Sulcus nasalis transversus — nasal root furrow

e

Sulcus anonymus superior — upper lid fold transition furrow

f

Sulcus orbito-palpebralis superior — upper eyelid furrow

g

Lineae canthi lateralis — lateral canthus lines

h

Sulcus orbito-palpebralis inferior — lower eyelid furrow

i

Sulcus anonymus inferior — canthal furrow

j

Lineae orbitales laterales — temporal orbital lines

k

Plica naso malaris — naso-malar fold

l

Sulcus orbitalis inferior — inferior orbital furrow

m

Sulcus oculo-malaris — oculo-malar furrow

n

Sulcus alaris superior — superior alar furrow

o

Sulcus alaris inferior — inferior alar furrow

q

Sulcus nasolabialis — nasolabial furrow

r

Sulcus nasooralis — philtral / naso-oral groove

s

Fovea buccalis — buccal dimple

t

Lineae supralabiales — supralabial lines

w

Sulcus angularis — mouth-angle furrow

x

Sulcus sublabialis — sublabial furrow

y

Sulcus mentolabialis — mentolabial furrow

ä

Fovea mentalis — mental dimple

ø

Lineae cervicales — horizontal neck lines

The legend names the labelled wrinkle, furrow, and line structures; it does not constitute a current scoring system.

Original format, export, and further processing

After acquisition, geometric surface data and photographic texture data were combined. The central point is that controlled views and different forms of representation can be derived from the same base material.

Show technical file formats

The original single scans are stored in the Konica Minolta/CDM format. Documented system output or export formats include STL, DXF, OBJ, ASCII points, and VRML; meshed 3D models were stored as VRML.

  • Konica Minolta / CDM (original single scans)
  • STL (*.stl)
  • DXF (*.dxf)
  • Wavefront OBJ (*.obj)
  • ASCII points
  • VRML (*.wrl; processed meshed models)

Protection of the material

The stimulus material is described only in general form. Person-related image data, individual case data, and raw data are not publicly displayed.