The reason we can see the color of objects is light rays undergo scattering in the subsurface of a material, and in their walk in the material lose some wavelengths, and finally exit at a random direction (link)

However, aside from this, you see white highlights on an object from the light source. These can be diffuse or specular depending on surface roughness (I'm assuming they're both from surface interactions)

enter image description here

Look at the highlights on the shoulder here for what I mean. Not where the arrow is pointing.

However, I just found that that aside from metal, most common surfaces, even when highly polished, specularly reflect only a few percent of the incident radiation. How are these highlights then so pronounced, especially for the diffuse highlight case where the light is distributed in all directions.

  • $\begingroup$ Humans can even see individual photons in the right conditions. Why would we not be able to see weak-but-visible specular reflections? $\endgroup$ Commented Nov 4, 2023 at 6:38
  • $\begingroup$ @HiddenBabel But they don't necessarily appear that weak compared to the intensity of the subsurface diffused reflection(i.e a few percent of the intensity). Even when taking a photograph, they are clearly visible in an ordinary exposure $\endgroup$
    – xasthor
    Commented Nov 4, 2023 at 7:07
  • $\begingroup$ I used outdoors as an example, but anywhere near a light source you'll see this $\endgroup$
    – xasthor
    Commented Nov 4, 2023 at 8:59
  • 1
    $\begingroup$ I guess one thing to consider is that the light from these two types of reflections is additive, so you're going to notice a localized spot that's brighter, even if the difference is not drastic. The other thing that the Wikipedia article you linked to says is that the amount of specular reflection can depend on the angle, so you may get a different percentage around creases and near the silhouette. Also, it's difficult to judge from images; in your photo, the person may be backlit (direct illumination coming from up and behind), and the front may be just light reflected off of the environment $\endgroup$ Commented Nov 4, 2023 at 12:47
  • $\begingroup$ @FilipMilovanović Indeed, in addition a photo may be re-touched with image editors like photoshop by blurring regions, changing contrast of image parts and etc. $\endgroup$ Commented Nov 4, 2023 at 13:21

1 Answer 1


What you are seeing in the photo is specular reflection from the surface of fibers in the shoulder area. That component of the light coming off of the shirt is white. Plus, the amount of such specular reflection increases with angle of incidence from the normal. You can convince yourself of this using a flashlight and a piece of fabric laid flat on a table, always keeping the angle of incidence = angle of reflection. For low angles, i.e. the light almost perpendicular to the surface, the amount of reflected light is much less than for glancing angles.

  • $\begingroup$ But something that is just a few percent of the incident radiation, even if increased somewhat by the right angle - how can it be so pronounced, especially considering the fact that on a rough surface it is being distributed in all directions $\endgroup$
    – xasthor
    Commented Nov 4, 2023 at 14:27
  • $\begingroup$ @xasthor - well, how do you know it only "increased somewhat" due to the angle? It could be that it increases dramatically at a certain threshold angle. $\endgroup$ Commented Nov 4, 2023 at 15:20
  • $\begingroup$ @FilipMilovanović Fair enough, that's possible. $\endgroup$
    – xasthor
    Commented Nov 4, 2023 at 15:36
  • $\begingroup$ @FilipMilovanović However, for reflection from a rough surface, light won't necessarily be reflected at that angle since the angle of incidence on some irregularity won't be the same as that it makes with the surface as a whole $\endgroup$
    – xasthor
    Commented Nov 4, 2023 at 16:44
  • $\begingroup$ @xasthor It's only "a few percent" at normal incidence but increases with angle. Look up Fresnel's Equations. For a rough surface such as a fabric on a table there are still many fibers lying flat or close to flat (depends on type of fabric). Try what I suggested with a flashlight keeping angle of incidence = angle of viewing (reflection). $\endgroup$ Commented Nov 4, 2023 at 16:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.