Post 23 - ICM Part 2 - The How and the Why
17-minute read, including photo captions
Just to recap last week’s post, I defined ICM (Intentional Camera Movement) as having these three aspects:
The camera is moving while the shutter is open. That movement can be vertical, horizontal, diagonal, rotational, subtle, or psychedelic – or all of the above.
The intent of the camera movement should be to create an abstract and emotive image, even if the original subject is still recognizable.
For these two reasons, Pan, Blur, and Zoom (alone) are not considered ICM, even though panning shots use the same motions as some ICMs.
The How
Now we turn our attention to how ICM is done.
Here’s the “Executive Summary”:
Lather. Rinse. Repeat.
I know….easy for me to say, right? Especially since I’m almost as bald as a baby’s butt. Nothing much to “lather” up there….
To a great extent, though, ICM is about playing the odds. I would estimate the “keeper-rate” at somewhere between 1 and 5%, depending on how you define “keeper”. That’s why continual experimentation and repetition is the key. So, if in doubt - try it out!
I wasn’t kidding in the last post when I said I had shot thousands of ICM images - take a look at a snippet of my folder structure below, and understand that this is just the tip of the iceberg.
To give you a more numerical perspective, in preparation for this blog series, I started sifting through my Lightroom catalog for ICM images and tagging them all with the “ICM” keyword. I was short-sighted in not adding keywords at the time when I imported those files, but at the time I write this, I’m about three-quarters of the way through the project of visually identifying all instances of ICM. Here’s what I have so far:
Total photos with ICM keyword = 7492
Rated as 3 stars = 317 (4.23%)
Rated as 4 stars = 74 (0.99%)
Rated as 5 stars = 7 (0.09%)
The star ratings are somewhat arbitrary and subjective based on when they were applied (and I haven’t even bothered listing the 1- and 2-star-rated, since I’m not sure they could be considered “keepers”). To be really accurate, I’d have to go through and re-rate everything based on my current knowledge and experience. But who’s got the time for that? I’m only showing you these numbers to make a point: you need patience when dealing with ICM.
Being organized - from the start - helps a lot too…. : -)
The Motion is the Potion
“Always in motion, the future is….” Yoda.
Although “hard rules” and “ICM” do not good bedfellows make, there are actually a few rough rules of thumb that have emerged for me over time.
Obviously, longer-than-normal shutter speeds are necessary, but there is such a thing as “too long”. A blurry, mushy mess is still a blurry, mushy mess - even to the abstract aficionado.
For bright/daylight scenes, I have found shutter speeds of 1/4 second to 1 second to be a good range (with ISO at the lowest setting on your camera, with the smallest aperture available on your lens). If you want to go longer than that in bright sunlight, you may need a 3-, 6-, or even 10-stop Neutral Density filter - or even just a Circular Polarizer to cut down on the light coming in to the sensor.
For morning/evening, blue hour/twilight scenes, shutter speed ranges of 1 second to 5 seconds can be appropriate - I’ve even gone as high as 10 seconds in some cases.
In general, the shorter the shutter speed, the more frenetic the motion - and vice versa (long shutter speed = more gentle motion).
Focal length also determines the speed of motion - a wider lens needs greater motion to produce a pleasing blur. A longer lens will accentuate motion, so make your movements more subtle the greater the focal length.
The motion used should correspond somewhat to the scene:
If you’re shooting something with predominantly vertical lines (like trees), a vertical motion is often the best to maintain a sense of the vertical linearity, as it were.
If you’re shooting a landscape/seascape or similar scene with horizontal separations of ground and sky (or other horizontal elements), then a horizontal motion is often better.
Sometimes a circular motion will produce pleasing results, especially for scenes with a mix of vertical and horizontal elements.
Subjects with strong points of light can often benefit from random squiggly motion, or sweeps, swoops, and teardrop motions.
Try these other movement options:
Random shaking and jiggling (up, down, sideways, diagonal - within the same shot)
Circular rotation (around an axis) and “wrist snaps” (be careful if you have a heavy DSLR!)
Even though I declared a zooming motion (when used alone) as non-ICM in my last post, don’t forget about zoom - it can produce dramatic results when used judiciously
Also try some shots out of focus, sometimes that can lend an interesting quality to an image.
Here are some examples:
Same building, same time, different motion. A vertical swipe (above) and horizontal swipe (below) emphasize different aspects of the structure. However, since the building has predominantly horizontal lines, the horizontal swipe shows the facade of the building better - you can (sort of) tell it’s a multi-story apartment building with balconies, windows, etc. In the vertical swipe, much of that detail is lost, and all you can really make out are the remnants of the vertical bars on the balcony railings. Of course the shadows of the trees in the vertical swipe add an interesting quality that is absent from the horizontal. So it really comes down to personal preference in terms of the level of abstraction and mood you want to convey - but the point is to be aware of the predominant shapes of your subject matter and adjust your motion accordingly.
A teardrop motion helps accentuate water droplets clinging to a bamboo grove, highlighted as the sun came out right after a rain shower.
The stronger, random/squiggly motion used for the wide angle shot above belies the calm, almost hazy, image that resulted. The relatively more subtle, vertical motion used in the zoomed-in shot below resulted in an interesting “separation” of the streams of the fountain, while still maintaining the shape of the grass in the foreground.
And don’t ignore zoom - it can be a very effective means of drawing the eye toward the center of an image when combined with a little shake and jiggle!
When your scene has a mix of horizontal, vertical, and other shapes, try combining different motions into the same shot, like this dogwood in blue-hour-bloom with a building in the background.
The Gear Has No Peer
Ok, I’m over this rhyming thing - this will be the last one, I promise.
In any case, you know how photographers often revert to the cliche “The best camera is the one you have with you”?
Yeah….No.
Even if somewhat overused, it’s not a bad sentiment in general, but in this case, gear is a crucial factor influencing the final image. You WILL get different looks with different hardware and software combinations.
This could be the longest and most complex section of this post - in fact, I could easily do an entire post on this subject alone - but I will refrain from that and just show you some examples to illustrate my point.
What it comes down to is the quality of the sensor - but not in the way you think. And here’s where my SECRET SAUCE comes in!
I promised in Part 1 that I would reveal a surprising secret - and here it is - drum roll, please…..
The Samsung Galaxy S5!
Yup, that’s right. I’ve got a high-end Nikon D810 “featuring a 36.3MP FX-format CMOS sensor without an Optical Low Pass Filter for particularly sharp and detailed still imaging” (to borrow a line from the official description). But that’s the problem. The sensor is TOO GOOD!
The Galaxy S5 camera has an older 16MP sensor that seems to fritz-out (that’s a technical term) in just the right way to make pleasing ICM images.
But wait - there’s more!
It’s not just the Galaxy S5 alone - it’s the combination of S5 plus the camera app “Camera FV5 Pro” that completes the recipe for the true secret sauce. (In other words, the native Samsung camera app on the S5 will not produce these results).
There is some, strange “je ne sais quoi” in that blissful marriage between the S5 and Camera FV5 that has produced many of the prodigy offspring in my ICM family. Which isn’t to say you can’t produce great ICM with a high-end DSLR. In fact many of the images you’ll see in the post and in the gallery were made with a DSLR - but there is a definite qualitative difference in the image produced by the S5/FV5 combination. It seems like something in the programming - the arcane algorithms - of the FV5 app that produces just the right kind of FRITZ (sorry to be so technical) in the S5 sensor.
Since I know what I’m looking for, I can spot the difference right away, but here are some examples:
DSLR: notice how the “good sensor” on my Nikon treats points of light as discrete lines (not a great photo, just using it to illustrate the point).
S5 with FV5 - although not exactly the same subject matter (points of light), it is the same motion, namely, a strong vertical swipe. You see how the blur has that sort of kaleidoscopic/psychedelic feel to it? That’s the difference I’m trying to highlight (but that’s very difficult to explain in words).
DSLR - this shot of a woman and her child riding down a slide at a carnival has a nice, pleasing blur to it, but….
S5 and FV5 - look how the “older sensor” and app deals with the lines and shapes of this shot taken from the back seat of a car while driving. Somehow, the S5’s sensor “stutters” more, and produces all those staggered lines. The DSLR would “blend” this scene into a smoother blur.
DSLR - a pleasing blur, but there are still discrete “lines” visible
S5 and FV5 - note how the cherry blossoms between the tree branches have that “painterly” feel, like they were done by “patting” or “daubing” a brush onto canvas (rather than “stroking” with the brush). Maybe that’s a good way to describe it: the DSLR is like stroking with a brush, and the S5/FV5 is more like plastering or spattering with a brush.
DSLR - tall grasses at sunrise
S5 and FV5 - tall grasses at sunset - not the same shot or vantage point (and very different motion), but I hope you can see the fundamental, qualitative difference between the two images.
DSLR - red blossoms in bright sunlight
S5 and FV5 - (different) red blossoms in sunlight - but can you see the difference in how these two technologies are rendering the same type of subject?
To further illustrate the point that “it’s all in the gear”, here’s another interesting hardware/software combination that produces yet another unique result:
Samsung Galaxy S8 with Cortex Cam
The Cortex Cam app is an image-stacking program, designed specifically for low light situations. You can set it to take up to 99 images, which it will then automatically align and stack to create virtually blur-free, handheld nighttime photos.
Here’s what it can do under normal conditions (when you’re not trying to “trick” it):
Not sure of the shutter speed for this one, but it was probably at least 5 to 8 seconds (handheld). Normally, an exposure that long would be a hopeless blur, but Cortex Cam was able to render the scene with a reasonable amount of stability (the lines on the buildings in the background are still quite visible and clean).
S8 with Cortex Cam - look at what this camera/app combination does when you start to shimmy and shake! This shot of a driver in a car has a unique and crazy “melting grid” pattern - due, I’m assuming, to the way that the software controls and interacts with the camera’s sensor under shaky conditions.
S8 with Cortex Cam - I honestly don’t even remember what this image was (I did a lot of random “point, shake, and shoot” while experimenting), but look at that wild, freaky grid-like structure. Could that be groups of pixels on the sensor? I don’t know, but it’s pretty cool, whatever it is.
S8 with Cortex Cam - I believe this may have been a video billboard or something like that (I was walking around town while playing with the app) - but I am just fascinated by the way the app has “interpreted” the data coming from the camera sensor.
S8 with Cortex Cam - I mean, try and wrap your brain around THAT, why don’t’cha! I think I should title this simply: “WTF!!!” Again, this is a single image, all in-camera, no overlays, no composites, no Photoshop filters, just some mad random shaking. Try getting that effect from a high-end DSLR/Mirrorless sensor - it ain’t gonna happen.
Alright, enough of that. The good news is:
You don’t have to buy expensive camera gear to create truly unique, fantastical abstract images. A (relatively cheap) phone-camera and app will produce amazing photos that will win friends and hypnotize people!
But all bets are off with other phone/app combinations - the S5+FV5 and S8+Cortex Cam are the only two combos I’ve tried. Who knows what other combos will produce - experiment and let me know what you come up with!
In case you’re wondering, the S8+FV5 did NOT produce the same kinds of images that the S5 did - there is something unique to the S5 sensor and the way it interacts with FV5 - I think maybe (ironically) the sensor was improved too much in the S8. (I have not tried the S5 with Cortex Cam).
You can still make very nice ICM images with your expensive gear. So your money wasn’t wasted. Whew!
The RAW and the Cooked
Or, in other words, post-processing. How do you edit an ICM image?
I happen to use Lightroom almost exclusively, although most of the other editing software available these days works in similar ways, so what I say below can be applied pretty universally.
Like with gear, I could wax philosophic about ICM post-processing techniques - but I’ll distill it to this:
Saturate and pop! That’s usually my preference. In my experience, colorful scenes with strong shapes and lines usually make the best subjects for ICM, so why not go nuts with the color? Push those sliders to the right! Play with the hues - sometimes changing the blues can have an interesting effect on the greens, and so on.
Soften or harden - experiment with Texture, Clarity, and Noise Reduction functions in Lightroom. The image below is an example:
The image on the left is the RAW file straight out of the camera - a Nikon D5300 (DSLR) - and notice how the lines and shapes are still relatively well-defined despite the blur. Adding considerable Vibrance and Saturation while also cranking both Texture and Clarity all the way down (to -100) produced the final image on the right.
To illustrate this point further, here is the same image (heading down the escalator in the Rosslyn Metro station) processed in three different ways - all three have been given the same Saturation, Vibrance, and other exposure settings. Only the Texture and Clarity have been varied:
Neutral - Texture and Clarity set to zero
Soft - Texture and Clarity set to -100
Hard - Texture and Clarity set to +100
Each image has a different “style” and “personality” to it - it’s entirely your choice which you prefer, but my point here is to (once again) experiment with all the tools at your disposal, both while shooting, and in post-processing.
Suitable Subject Matter
Almost anything is appropriate for ICM treatment. When I first “discovered” this technique, I was mostly drawn to bright, contrasty, colorful scenery, but as I have seen more of the work others have been producing, I’ve gained an appreciation for the “subtler approach”. In the future, I will be trying some B&W ICM, some layered composites, and scenes with more muted colors.
Just to give you an idea of the breadth of subjects I’ve shot, here are some examples:
Carnivals
Carnivals and state fairs are a great venue for ICM since there is already so much color and motion going on.
Merry-go-round with horizontal swipe
The life of the carny - hurry up and wait. The boredom of this idle kiddie-ride attendant is evident even through the blur.
People and crowds can make great subjects - I loved the gentleman’s pink turban and the way it contrasted with his bushy white beard, but he was walking past so I shot this from the hip without really looking. I didn’t even know if I got him in the frame until I looked at the photo later. (Shot at the same carnival as above).
Keep the motion going - this was shot out of the windshield of my car while stopped at a traffic light. The oscilloscope-lines that almost look like a sine-wave were made by a car passing in front of me, but instead of a single vertical swipe, I kept the vertical motion continuous while the shutter was open (1.3 seconds).
Warp factor 5….ENGAGE…. A driving shot, camera perched on steering wheel.
Don’t forget to try some selfies! This was done with the S5+FV5 - and another example of why repetition is key. I took 104 shots in this series, but this is the only one I liked enough to use. In fact, I made it my Instagram profile pic. I’ve also used it in various other places, including my first blog post and Contact page. (I know, it’s a bit creepy, but I like it - what can I say)?
TV captures
This is a subject matter that I’ve seen only a few photographers explore, but I absolutely love it!
All it takes is sitting in front of the TV, waiting for colorful scenes in shows or movies, then firing off a few jiggling abstracts when what’s on the screen looks interesting. All of the examples below were shot with either the Galaxy S5 or S8 since phones are lighter and easier to hold up for long periods of time - and they produce better “abstractions”.
S5+FV5
S5+FV5
S5+FV5
S5+FV5
S5+FV5
S8 + Cortex Cam
S8 + Cortex Cam - If you know your sci-fi, you may recognize the film this scene comes from….but this also demonstrates another reason why TV captures can be interesting. This happens to be a transition from one scene to the next, hence the inclusion of two faces from different cuts.
S8 + Cortex Cam
S8 + Cortex Cam
The Why
I’ve already hinted at the why, so the answer shouldn’t be too surprising: abstractionism (is that a thing?). Ok, impressionism - that’s a thing, right? Maybe expressionism is more appropriate, since that often seems to be more about creating emotional responses with vivid colors. Who knows - as long as there’s some sort of “ism” involved for the sake intellectual clout, I’m cool with it.
Ever since I was young, I’ve had an affinity for abstract art, impressionist painting, etc. Van Gogh is one of my favorite painters, and I’ve been fortunate enough to go to the Van Gogh museum in Amsterdam several times. But I also love the work of Monet, Manet, Renoir, Degas, Cezanne, Matisse, Seurat, Hopper, and many others.
When I learned about Dadaism later in college, it opened a whole new world of absurdist art, which had a huge influence on my artistic preferences.
Then came personal computers - and a whole new art form with them. One of the first “digital artists” whose work I was exposed to was Laurence M. Gartel - specifically, his book entitled “A Cybernetic Romance” (published in 1989, but I didn’t come across it until about 8 or 9 years later). In that book, there was a piece called “The Immaculate Queen of Technology” that has become one of my most favored works of art to this day. I even tried (poorly) to recreate the piece as a painting at one point. Here it is:
The original “Immaculate Queen of Technology” - scanned from the book
The painting I did (from a tracing, then the head was cut away and glued onto foam core for a “3-D” effect)
It’s no accident that this image shares some basic similarities with some of the directions I went into with my ICM images. In fact, the image below, which I featured in last week’s post, vaguely reminds me of “The Immaculate Queen” (it’s an S5+FV5 TV capture, by the way):
Many years later, in February 2006, I was fooling around with my first digital camera, the newly acquired Canon Powershot S2. I happened to be sitting in front of the TV, and accidentally fired off the shot below. I didn’t even notice it at the time, but later when I downloaded all the “test” shots from the memory card, this one “spoke” to me for reasons I couldn’t really explain. To me it looked like an abstract version of “the angel of death” (or something). It looked to me like it was wearing a black cape, swinging a sword across the frame. That red line was a spray of blood coming toward me. Ok, ok - totally morbid, I know - and I’m really stretching here, but this image stayed with me for whatever reason. You can put me on the couch and psychoanalyze me later. It didn’t really remind me of “The Immaculate Queen of Technology” visually, but it had the same impact - and import - in my visual cortex.
I’d say this was the first real ICM image I made - although I didn’t realize it at the time.
Fast-forward 11 years, almost to the day (02-02-2006 to 02-04-2017), I was standing on my balcony after sunset, and again - quite accidentally - took this blurry shot. Looking at the back of the phone, I thought to myself, “It’s always the mistakes that look the coolest, isn’t it”? This was taken with my Samsung S5 with the native camera app (I hadn’t found the Camera FV5 app yet at that point).
Being the stunted dullard that I am, I basically shrugged off that first “sunset blur”. It wasn’t until two weeks later (02-17-2017) that I was trying out the freshly-downloaded Camera FV5 app. I was sitting in front of my MacBook Pro, which has a screensaver with all the album covers in my iTunes library, shown in little squares that change out individually at random. This was another “accident”, although it was sort of an “accident-on-purpose” in the sense that I was trying out the longer exposures possible with the FV5 app, and waving the camera around without any real goal in mind. It was at this point that it FINALLY dawned on me that maybe purposeful blurring could be a “thing”. I got more and more excited the more I tried it out, but it wasn’t until months later that I found on Google that there were many others doing the same thing, and they had a name for it: Intentional Camera Movement.
And the rest, as they say, is history!
There are more photos not shown in this post in the Gallery - don’t forget to check that out. Enjoy!
As always, thanks for reading my blog posts!
Next week we’ll take a break from ICM and go for a walk in a park! Time to unlock the apartment door and poke our heads outside! How’s that sound?
Share! Feel free to share this page with your friends! Copy and paste the link, or use one of the buttons below:
https://www.cameralingua.com/blog/post-23-icm-how-and-why
Thanks again!
To search the site, start typing any keyword….
