I can’t draw
24 February 2008
One of the many reasons why I take photographs is because I can’t draw. Well, I can draw technical drawings, within limits, but it takes a while, and the human form isn’t amenable to being drawn that way. So I take photographs. But I like sketches, and I’d like to be able to sketch. Now I’ve found an interesting option.
PhotoShop has an array of filters that do strange things. I doubt I’ll even scratch the surface of the filters built into CS3, let alone indulge in the multitude of filters that are available for download, either for money or for free. That doesn’t mean I don’t play around sometimes. I had fun playing with the Emboss filter a while back. Doubt I’ll ever use it for real, but it does make some cute images. Maybe if I wanted to turn an image into a button for a windows interface of some kind.
The one filter I use on every image is Sharpen – I generally use Smart Sharpen, because I prefer the results I get with it. Old-style PS users tend to stick to Unsharp Mask (never have understood why it’s called “un-sharp”, but that’s not important).
Anyway, I occasionally explore the Filters menu when I’m playing with PhotoShop. A few days ago I tried using the Find Edges filter on an image I’d converted to a semi-silhouette (I talked about semi-silhouettes a while back). The results surprised me. I got an image that looked to me exactly like it had been sketched using coloured pencils. Not necessarily the right coloured pencils, but coloured nonetheless. That is rather neat! If I converted the result to monochrome I could end up with a regular pencil sketch, but I like the coloured version. This is very exciting for someone who can’t sketch, but wants to. In the example semi-silhouette here I think it’s really interesting how the sketch shows the face as blank (because it was black in the original) – it looks as though the sketch is unfinished, as though the artist left off before sketching the face. That adds to the feeling of the image being a sketch, for me.
Having made this serendipitous discovery, I started trying it on other images. It rapidly became clear that my fondness for clean white (or black) backgrounds was a huge plus when using this technique, because it yielded a clean background – another thing that enhanced the feeling on this image being a sketch. Using the silhouette technique first isn’t a requirement (although it does produce strong edges, which doesn’t hurt). The next example isn’t a silhouette – it’s the same model, coated in olive oil (it’s good for the skin, apparently, as well as making her smell like Greek food – amusingly, we had treated ourselves to Greek tavern food before this shoot).
The background doesn’t have to be white – it can be black (I discussed the special features of white and black backgrounds another time). It is easy to make completely pure black and pure white backgrounds. That really helps when you want to produce these sketch-like images, because it means there are no edges in the background. If we use strong edge lighting we can get thick edges. This image was lit using two large slot softboxes, positioned left and right of the camera, at 90 degrees to the camera angle. This results in an image that looks like it was outlined with a blunt 6B pencil. I toned the edge down a little by using a large round brush painting pure white (another good reason to get a clean white background).
This is one of those effects you can over-use (although not as quickly as Emboss). I won’t process every image I take this way – that would be ridiculous. But it produces attractive results that mimic sketches. Suddenly I can make a kind of art I have never been able to make before, and that means a lot to me – hence this write-up.
I think I’ll go play with some more of the filters in the PhotoShop menu…
Implied nudes – the tacky nude?
2 February 2008
Implied nude is an important variation of nude photography. Implied nudes can be displayed in places that would be inappropriate for “real” nudes, so implied nude images can be more useful. That’s not the only reason to be interested in implied nudes. Sadly, they can be seen as a tacky alternative to real nude images. This is probably the legacy of decades of poor quality implied nude images on the front covers of cheap magazines, and the exploitation of implied nudes in advertising.
A good implied nude can be more erotic than a bad “full” nude. I think of images as expressing a statement. A bad nude image says something like “me girl – see my girly bits!” – a neanderthal lack of subtlety. A good implied image says something like “I seem to have mislaid my clothes. If I were to move a bit you might see more than you would normally. What a pity you can’t see it here” – it is a tease. It’s the difference between hitting the viewer over the head with naked breasts (an interesting image – I must consider shooting that some time), and making them imagine what is hidden. I think that’s fundamental to a good implied nude – the image must be teasing the viewer. A good implied nude stimulates the imagination; it can also be more erotically stimulating – that’s not a bad thing. Indeed, one might argue that implied nudes are more about eroticism than artistic nudes – a really good artistic nude can desexualise the image to allow the viewer to concentrate on the shapes – that doesn’t happen in an implied nude. Hiding something draws the viewer’s mind to it; exposing it removes the mystery.
There are lots of bad implied nude images – these are the ones I was referring on the covers of cheap magazines. Images of a woman covering her breasts with her hands – what are these saying? If she didn’t want to be photographed topless, why did she pose that way? The coyness feels so fake. The same thing applies, to a lesser extent, to images where she is holding a towel or sheet to her breasts. These shots lack imagination. They taint the whole genre.
I like bold implied nude images. Images that are only barely implied (pun not intended, but I like it!). Images that look, at first glance, as though they are revealing everything; look more closely, and you realise that the crucial areas are covered (even in the original high resolution image, nothing shows). Look at the dining chair image – the young lady in question is clearly nude, and not sitting in a discreet way, yet she is showing less than she might in a bikini (well, a high-cut bikini…). This is not a subtle image. Some might take exception to it, despite the fact that it is not explicit, but then some take exception to the whole idea of taking photographs of unclothed people – makes me wonder if they are objecting to being reminded of their own sexual nature?
Now compare that image to the classic implied nude with the heart-shaped cushion. This is more like the common idea of implied nude imagery. There is nothing wrong with this image, and I like it, but I think it lacks the impact of the dining chair image. That gave me cause to ponder for a while. Is it that the cushion is an extraneous element? Could it be that the chair is standing symbolically for a person, and the pose suggests her body wrapped around her lover – most emphatically a sexual image? Or could it be as simple as the image showing her legs apart; her legs are demurely together in the heart image? Exploring the ramifications of that question is a topic for another day.
I think we can extend the idea of the implied nude to images that show breasts – I’m just not sure what to call them. The young lady sitting cross-legged is an example. One reason I consider this an implied nude is because the pose suggests that the viewer might see more if the point of view were swung around the left. You might argue that this is what some refer to as “tasteful” nudes – I don’t like that term, because I think it is possible for an explicit nude to be tasteful – consider Michangelo’s David. I don’t like calling this “implied nude with breasts”, either. Perhaps we need to invent a new term for it? Something like “discreet nude”? I don’t see this as a purely artistic nude, because of the implied sexual nature of the image. There are artistic elements, such as the strong curve to her spine, and the deliberate echo of the red of her lipstick and the stool, but it cannot escape the sexual insinuation. That is not a negative, it is just a distinction, of the kind that is involved in defining categories – defining categories is the splitting of increasingly fine hairs.
Perhaps I shoot implied nudes in an effort to liberate the implied nude from the tacky associations it has acquired? Nah, that’s not it. I think I shoot them because the models want them, because there’s skill involved in arranging a shot so that some things are hidden and others aren’t, and because I like the resulting images. Maybe I’m a tease at heart? I can think of much worse things to be…
Some background on studio backgrounds
4 November 2007
When we are shooting outdoors, we have backgrounds provided for us. Oh, we pick and choose, asking the subject of the photo to move so that we can capture them against the better background, but generally speaking we have a limited choice. If we are shooting indoors, but on location, once again, our backgrounds are pretty much dictated by our environment.
Not so in the studio. Here we are in control. If we wanted a candy-striped background we can have it (I may regret saying that…). We can even shoot against a chromakey background, and swap in any background image we want. Despite that, you’ll see three colours of background in the vast majority of my studio shots: white, black, and grey. There are good reasons for this.
When I am in a studio, I want attention focussed on the beautiful woman I am shooting. Using a simple background helps – there is little to look at other than the model. I’ve heard some people refer to a white background as “catalogue shooting” – the same techniques are used when shooting items for inclusion in a catalogue.
Black and white backgrounds are easy, at least in principle. When we bring a RAW image into Photoshop, we get to specify the levels that are considered ”black” and “white”. To get a pure black background we ensure that all of the background falls below the “black” level. To get a pure white background we ensure that all of the background is above the “white” level. This only works for black and white – I can’t force a blue background buying having the background fall above or below a specific level. If the background falls between the “black” and “white” levels, it will come out as a shade of whatever colour or pattern it happens to be.
Black background
To get a pure black background, we want the background to reflect much less light than the subject. We start by moving the subject away from the background, and arranging our lights so we spill as little light onto the background as possible. The less light we get on the background, the easier it will be to make it pure black. (in the image, Phoenix is sitting on a satin finish surface, hence the reflection below her)
It helps to be using a dark (ideally black) backdrop. I’ve used a variety of offerings, including a synthetic velour, which worked well enough, but was a magnet for lint and dust. My current favourite is a fabric that can be obtained from a theatrical supplier – its technical name is black Molton, but it is known in the trade as “commando cloth” or “blackout curtaining”. It’s a black brushed cotton fabric, pre-treated with flame-retardant chemicals. It comes around 3m wide, which is plenty for my home studio, but it is easily made into any desired width – it is used for the black curtains that go behind the fancy one in a theatre, after all – and it soaks up light. It is possible to get this fabric to reflect light, but you have to work at it!
White background
A pure white background is a little more complicated. This time we want to pour the light onto the background, instead of keeping it away. I’ve found the easiest means is to dedicate two lights to illuminating the backdrop, positioning one each side, aimed about one third of the way in to the centre. Conventional wisdom has it that you light the backdrop 3 stops brighter than the subject to make it white. I have experimented, and I rather like the results I get when I run the back lights at full power (2 x 600Ws for my current lights) - the extra light wraps around the model, giving an attractive rim-lighting, and a form of silhouette, if I don’t put a light on the model directly. Tavia is demonstrating this effect.
It helps if the backdrop reflects light well (although you can get almost anything to look white with enough light – I have turned a dark brown surface into white with a LOT of light). A plain white backdrop is good – I have used white polyester, white velour, and white paper – they all work, although a smooth surface can show irritating highlights, and a patterned surface is a disaster if your background lights aren’t bright enough. I have not been able to get white Molton, even though I’m assured that it exists – there just isn’t enough call for it (unlike black Molton). I am currently using a fabric from the same theatrical fabric source – this one is intended for use as a reflective surface, and it works well – it appears to reflect about a half-stop more light than the white polyester I was using previously. The only drawback is that it has a textured surface – if it is not blasted through to white, there is a texture to deal with.
Bear in mind that having the background very bright will upset Camera Raw’s normally reasonable auto processing. That’s OK – we just need to do some unusual processing.
Camera RAW processing
Once we have the RAW image, we need to bring it into Photoshop appropriately. The instructions below refer to Adobe Camera RAW for Photoshop CS2. The instructions for CS3 are similar.
If we want a black background, we need to raise the Shadow slider until the background is completely clipped to black – hold down the Alt key (on Windows) or Command (on Mac) while moving the slider to see the clipping.
If we want a white background, we need to work with the Exposure slider. We will be clipping the background again, but this time to white.
If we are shooting a full-length shot, it can be quite difficult to get the floor under the model to clip – usually we need to compromise. I have had some success using a narrow slot box to illuminate the floor when shooting for a white background, but it’s a work-in-progress. This isn’t a problem for shots that don’t go down to the floor.
Essential shadow
22 April 2007
The art of photography is about light and shadow. The light is obvious – try to take a photograph without light, and you may have difficulty interesting people in your artfully posed black cat asleep on a black rug in a black room at midnight on a moonless night… However, shadow is at least as important as light.
I learned this lesson when I started taking photographs. I carefully lit my subject, producing a very even light (something I now know is called “flat lighting“), and an amazingly boring photograph; it was flat and lifeless. Thus began my adventure to learn about chiaroscuro (not that I knew that was what I was looking for). I hadn’t studied art, so I was oblivious to many centuries of learning on this subject.
You see, despite their complete lack of photographic equipment, painters of years past managed to produce images that were not boring, flat, and lifeless (well, some painters did…). They did this by painting not just light, but shadow too. We can learn from this.
A photograph or a painting is a two-dimensional object, but we are trying to capture three-dimensional space. In flat lighting we get a two-dimensional image that looks completely two-dimensional (one day I will shoot that deliberately, but I need to compose the idea carefully). There is no depth to the image. Such an image is often emotionally unsatisfying to the viewer. We expect to see depth, and we see it in the shadows. We can tell that an object has three dimensions because of the shadows cast on one part by shadows from another. A flat lighting setup eliminates those shadows, and eliminates the sense of depth that they conjure.
The way we achieve our necessary shadows is to use unbalanced lighting – if we use a bright light to one side of our victim (er, subject), it will cast shadows on the other side. If we place a bright light above, it will cast shadows below, and vice versa. Using a single bright light, we may get shadows which are too intense. We can reduce the intensity of the shadows by placing a second light to throw light into them. If that second light is as bright as the first, and placed at the same distance, it will completely destroy our shadows. That’s no good. So we make the second light dimmer, and/or further away. One ratio of brightness that has been shown to be effective over years of photography is having 3 stops of difference between the bright light and the dimmer one. That’s not to say that we shouldn’t try other ratios, just that 3 stops is a good place to start.
Photographers who still use meters will trigger one of the lights and measure the result, then trigger the other, measure the result, then adjust the lights accordingly. I don’t. I position the two lights at the same distance from the model, and at the same angle (typically at about 45 degrees from the line between the model and the camera), and set them about 3 stops different – the lights display their intensity in (sort-of) f-stops. The Elinchrom 300RX ranges from 0.5 to 5.5, and setting 1.5 on one and 4.5 on the other yields a difference of 3 f-stops.
I wonder if anyone has analysed the great masters of portrait painting, and determined what the effective f-stop difference is between the two sides of the faces they painted?
Playing with perspective
28 March 2007
Any time we use a lens that has a focal length different from 50mm (on a 35mm SLR) we produce images with a different perspective from the human eye. A lot of the time, perhaps most of the time, we want to avoid making that obvious. There are exceptions – when we play with a fisheye lens, it is rather obvious that we’re not seeing the image that we’d see if we placed where the camera was placed to take the shot.
When we use a telephoto lens (one with a focal length greater than 50mm), we compress perspective. Things that would look further apart to the human eye look closer together viewed with the telephoto lens. When we use a wide-angle lens, we exaggerate perspective – we make things look further apart than they would to the human eye. We can play with that effect.
If you’ve done any reading about photography you may have read warnings against using a wide-angle lens to shoot portraits. “Don’t do it!” the books warn, “It will distort features, it will make people look like they have huge noses. The results will be horrible!“. Some books will even show you examples, and they do look distorted.
However, that’s not all of the truth. Sometimes it can be very effective to use a distorted perspective. After all, we’re not making a documentary – we’re making art. Art can distort things to raise questions about them. Or to throw something into relief – to make it stand out from the background – to draw attention to it. Or we may just be trying to flatter our model: we may be trying to make her look as though her legs are longer.
In anime (Japanese animation) there is a technique called “super-deformed” characters, sometimes also called “chibi”. To indicate that a character is having a tantrum (or is very emotional), the animator may draw the character with a very large head and small body and limbs. This makes the character look cute, and generally suggests that we shouldn’t be taking the tantrum too seriously.
I wanted to get a similar effect with a model. I struck a problem, in that I can’t rub out a model’s head and redraw it – I’m just not that good an artist. So I switched to a wide-angle lens and did all the things the book tell us not to do. I zoomed to the most wide-angle setting (24mm on a 24-70mm f2.8L), then I got close to the model to maximise the ratio between distance from lens to face, and distance from lens to feet. That’s how we exaggerate proportions.
You can see the results in these images. I don’t think it’s perfect, but it worked surprisingly well for a first try. I’d be interested to hear what you think of this technique.
