Orio
Offline
Image Upload: Off
|
 p.18 #25 · Where does the 3D look come from? | |
I have personally begun to realize a few environmental conditions that help a lot the perception of roundness:
1) Light direction, the more lateral the usually better - although I have seen very "round" images even with frontal lighting
2) having different "wings" in the framed scene, at different distances
3) the "aerial perspective" of Leonardo heritage (haze in the background distance)
4) Having round or cylindrical objects in the frame (e.g: a tree trunk; a human face; etc.)improves the perception of depth for the whole image; Cubic objects, even of greater physical depth, help this perception less. Other wise said, a small trunk or human face can help more than a huge wall, ever if the wall is at slanted position in the frame. Of course this is also related to the points "a" and "d" (more below)
There are of course photographers choices and conditions that are equally and probably even more important:
a) with regards to point 2 above, it is important to have a lens that has a smooth gradual transition from focused to blurred. The smoother the transition, the greater the depth perception
b) Having a photo with a low macrocontrast and a high microcontrast not only helps a lot, but is also a key element apparently
c) provided that the other conditions are there (especially point "a"), some wisely applied sharpening does enhance the effect of the microcontrast and thus boost the depth perception. it is important, however, that the sharpness is not so strong to also bring out too much detail from areas that are placed in a different scene wing. This also works in relation to the following point "d"
d) the DOF is a very important choice to make. And of course it varies with the lens you use, so it's difficult to make an overall rule for that. Based on my experience, however, I'd say that neither extremes (wide open or too narrow) work well. When wide open, you end up flattening the perspective too much, and make things visually end on one same "image plane" where they would actually be separated by good distance.
When stopped down too much, you also have a similar consequence of flattening, because having too many objects at different distances in perfect focus, also flattens the perspective. Opposite action, but same consequence.
The best "3D" results seem to be obtained for me at moderate apertures - I would say in the range from f/2.8 to f/5.6
e) White balance: with regards to point 3 especially, I have found out that warmer tones tend to reduce the depth perception, while cooler tones tend to enhance the depth perception. This has of course to do with the fact that the air actually has the blueish colour of the sky, so to have a cooler tint to the image means to reinforce the impression that there is more air amongst the objects.
This of course does have to be subtle, because if the cool cast is too heavy, it will have the opposite consequence of flattening the perception of the space, because the stronger the cast, the more it reduces the number of different colours in the image. This is also related to the following point "f"
f) Having a greater number of different colours in the image seems to enhance the perception of depth, although slightly. Of course the placement of colours also matters: having warmer colours in the front and cooler colours in the background helps to keep things within the Leonardo aerial perspective rule
g) Exposure: based on my experience, I have noticed that slightly underexposed images render more perception of depth for the foreground objects especially, and in general for the focused objects. I have not thought about this much to analyze the reason why - yet it's something that I can visually appreciate in my tentatives. Perhaps this is linked much to point "b", as it seems that the more microcontrast a lens can render, the more pronounced is the depth effect of the underexposure of the focused subject and foreground.
h) ISO, the lower the better. I find that image noise or grain can largely kill even the best achieved depth perception result.
OK these are more or less the results of several weeks of tentatives and reasoning. Maybe there is something else that escapes my memory now, but mostly i'd say that this is it for the current state of my research.
|
