Metric fitting results
Metric "HDR-VDP-2.0"Test set "CSF flattening"
R = 2.4 dB
chi_2_red = 2.1

See the gallery of stimuli.
About the data set "CSF flattening"
CSF flattening due to the masking signal
This data set was measured to demonstrate the effect of CSF
flattening in complex images. Gabor patches from 4 to 16 cpd were
superimposed on an actual image (portrait) in three different regions:
the region with almost no masking (hair), with moderate masking (face)
and with strong masking (band). The hair region has almost no masking
because the pixel values are clamped at 0. Note that also the
luminance varies greatly between these three regions.
The thresholds were measured in the 4-alternative-forced-choice
experiment, in which all four images were shown side-by-side. The
measurement procedure was the same as for the "Contrast sensitivity
for wide luminance range" data set.
About the metric "HDR-VDP-2.0"
This is the proposed metric described in
detail in the paper "HDR-VDP-2: A calibrated visual metric for
visibility and quality predictions in all luminance conditions" (doi). It shares
many similarities with VDP'93 and HDR-VDP, as it was inspired by these
metrics, but the functionality is much extended and individual
components are thoroughly revised. The major differences are:
- The metric predicts both visibility (detection/decrimination)
and image quality (mean-opinion-score).
- The metric is based on new CSF measurements, made in the
consistent viewing conditions for a large range of luminance and
frequency.
- The new metric models L-, M-, S- and rod sensitivities and is
sensitive to different spectral characteristic of the incoming
light.
- Photoreceptor light sensitivity is modelled separately for cones
and rods, though L and M cones share the same characteristic.
- The intra-ocular light scatter function (glare) has been fitted
to the experimental data.
- The model used a steerable pyramid rather than cortex transform
to decompose image into spatially- and orientation-selective
bands. Steerable filter introduces less ringing and in general case
is computationally more efficient.
- The new model of contrast masking introduces inter-band masking
and the effect of CSF flattening.
- A simple spatial-integration formula using probability summation
is used to account for the effect of stimuli size.