Convex programming colour constancy with a diagonal-offset model

Robust Tracking for Augmented Reality

“…Unfortunately these conditions are not fulfilled in our experiments and, consequently, the diagonal transform cannot successfully map the image gamut into the canonical gamut. Thus, we have selected another model, the diagonal-offset model (Finlayson, Hordley, & Xu, 2005), which considers six parameters to map the image gamut. Let be a color of the canonical gamut and an observed color of the image gamut, then the diagonal-offset model maps this color using the next equation:…”

“…These solutions can be discarded by computing the distance between the observed values and their expected canonical values, and removing those that are farther. And finally, as it is recommended in (Finlayson, Hordley, & Xu, 2005), we should look up among the remaining feasible solutions for the solution with a lowest offset vector. A worse situation is shown in Figure 4.…”

Evaluation of color descriptors for object and scene recognition

“…For example, a shift in the color values due to increased 'diffuse' light cannot be modelled. To overcome this, Finlayson et al [7] extended the diagonal model with an offset, resulting in the diagonal-offset model:…”

Color Image Registration under Illumination Changes

“…There are other cases where illumination compensation can fail, for instance, if there are other processes happening like bias in the camera, or saturated colours in the scene. In the latter case, some colours would fall in (and outside of) the camera gamut boundary [4]. This is the reason why the use of a complete (full) affine transform in the form Ω · I(q) + Φ is justified (see, for example, [5], [8], [16], to cite a few), where Ω ∈ 3×3 is a full matrix, with elements ω kl , k, l = 1, .…”