Post Processing Briefly Noted
by Bob Weil
I am delighted to announce that iPC's Bob Weil will be sharing his post- processing skills with us on this page on a regular basis. If you need a fairly regular fix of new ideas and techniques on iPhoneography, may I suggest you stop by here often!
Over to you, Bob!
Thanks, Nicki! As I have a chance, I'll drop images into this column with a few words of explanation about what tools I used to post-process them. Sometimes I'll just mention the apps and the some of the technical challenges I had, and at other times I'll provide a little more detail. I hope everyone finds them interesting! Be sure to scroll down to see older entries.
December 29, 2012
Creating a surreal collage in 5 hours, over two days, in 104 steps
Using Jazz!, Photogene2, XNView and Superimpose (a lot)
Tiny Planets ~ (the contorted building in the first frames)
Pixlr-o-matic ~ (colorizing and adding a file to a local building under construction)
Jazz! ~ (colorizing and re-colorizing multiple versions)
XNView ~ (colorizing and re-colorizing multiple versions)
Superimpose ~ (layering of new elements, new backgrounds - often painted wall or floor surfaces - and bringing back the color of the fire/flame as it was "extinguished" through repeated processing. But just as importantly, re-layering of previous versions to change the coloring, tone and mood and to add flatness to the image)
I was not feeling well across the long holiday weekend, and between medications and holiday wine infusions, I felt the need to try to do something constructive as I was laid up in bed. I found an interesting background I had created using Tiny Planets (from a high rise office building going up next to ours) months ago, and decided on Christmas day to create something from it. I had no idea how it would evolve. I went on an extensive tour of my camera roll, going back two years, and found 8 images that couldn’t be more diverse.
I decided I would figure out how to pull all of them together as I went along. In three instances (the woman, the hand and the powerblade) I had to go back to the camera roll to fill in “gaps” in the story. Then, I blended several background textures (photographed crumbled wrapping paper, scratched metal, and a borrowed bit of wonderful looking script for the “love letter” reference in the upper right). A great deal of the depth in my pieces comes from the systematic layering of different images (walls, painted floors, etc.) using different blend modes and opacity settings.
So, below, you’ll find 104 images, chronicling each intermediate “save” I made, starting with Photogene2, and working through Superimpose to Jazz! to Camera+ to XNView, and then back to Superimpose (repeatedly).
What sorts of ideas did I set hope to convey? Well, to be honest, they accumulated as I added new elements to the image. Love, gained and lost, the fire of passion, burning fiercely and extinguished. Seeing things in hindsight - and imagining what they might be in the future (the child in the lower left carrying a miniature version of the building that passion built - her future?). The crossed-out lines from a letter on the left, the hand-written words in French, reading: "The Love Letter to my Neighbor". The hacksaw blade, symbolizing the tortures of love - the raven, the end that awaits us all. The beautiful figure at right - the graces and rewards of love.
The final image:
(I’m not sure why, but in outputting these in series, the sequence for each group of nine runs from the lower left to the upper right, as shown in the first image below.)
September 13, 2012
Creating Drama Where There is None
Using iDesign, Laminar, iColorama and Superimpose
Conceptual photography is more like narrative filmmaking than it is like street photography – the goal is to express a story conceived in the mind of the photographer and not necessarily to reproduce what the lens sees. And the process may more closely resemble painting if the original image is only the substrate for a series of composited layers that completely redefine the scene.
In my case, I often look at images in my Camera Roll to see where a story may arise by inference (post processing decisions, image titling, etc.) or by reconstruction from the ground up (compositing or collage of new elements into a scene).
I was recently an observer at a photoshoot, and the photographer and his assistant were taking a break between shots. I captured them as they chatted about the next set up. With no editing, it’s probably a shot I would have discarded. But I saw an opportunity for something a bit more intriguing than the scene actually represented in “real life.”
In the Before and After images below, you can see that by cropping closely in on the subjects, I create the impression something more profound than a simple workplace conversation was occurring. The careworn expression on the woman’s face and the relaxed look on the man’s face clamored for a backstory.
So I cropped the image in such a way that neither face was shown in full. This automatically, without any additional steps, creates tension in the eye of the viewer. It’s a bit unnatural, and the viewer is forced to concentrate on the facial expressions of the subjects and not the overall scene or the immediate environment. Then I added elements that reinforced the idea of two people in conflict - a line dividing them, color differences on the two sides, his face half-toned, and the words (angry conversation?) spilling over the scene.
These elements in concert with a grunge texture and color caste suddenly lent meaning to the differing facial expressions - his offhand look and the fact that she is staring fixedly into the lens convey powerful new emotions that weren’t actually present.
After cropping the image to the size and proportions I wanted (I really try to make most of images a perfect square) in Photogene2, I bring it into iColorama to process a version of the image using the Halftone setting. My purpose is to merge the original and the halftoned version so that only the man's face is halftoned.
In a later step not show here, I mask and merge the original without halftoning and the copy with halftoning - using Superimpose to leave him halftoned and return her to the original look.
I then bring the image into iDesign to add two graphical elements – a vertically angled dotted line and a colored semi-transparent overlay for one side of the image. To complete the image, I add a splash of letters and word fragments, a lomo color caste and grunge texturing.
iDesign looks nothing like any computer-based program you’ve ever seen, but it’s logically laid out – and if you’ve used vector programs before (that allow you to create lines, shapes and type that are resolution independent), you’ll catch on right away. One amazing feature is the ability to export a PDF as well as the standard jpg or lossless PNG. You start with an empty canvas and import the image as a background, then you’re able to add elements in layers over top of the image.
Below, you can see some of the options open to you when you are creating an object – you can create a line, a box, a circle or oval, or a text area. I created a line, and I was presented with the option to set width, color, solid or broken in a variety of styles, etc.
After creating the line, I color it to a percentage gray:
I then use the same set of tools to create a box, and then change the color to a tan while removing the default outline:
I then output the image as a PNG, and process it using iColorama to create a half-tone effect. In another step not shown here, I import the original and half-toned versions into Superimpose, where I use masking tools to blend the two versions using a gradient - leaving his face half-toned and returning her face to the original look. This further reinforces the fact that she is the main subject.
Also in Superimpose, I add a layer of letterforms haphazardly scattered across the page. The text here implies unintelligible communications between the couple – it’s a way of representing noise visually.
I love Superimpose because it is one of the layering programs that has the most layering modes, and it allows you to see a preview of the blend at the selected percentage without first applying it. It also has fairly powerful masking tools, second only to Photoshop Touch. The only downside – you can only have one layer at a time.
Then I bring the image into Laminar, one of my favorite programs for managing multiple layers at once. It has a fairly high number of blend modes and excellent (although quirky and sometimes buggy) masking and selection tools. And it manages up to 16 layers simultaneously (hurray!).
I first add a spotlight effect to highlight her face and draw the eye to her expression and clearly make her the primary subject of the image:
I then use the FX set of controls to add several textures and color modes – Laminar has a very usable, although small, range of presets that can be adjusted after application. They can be applied directly to the layer, or applied as a new layer (duplicating the content of the layer you’re operating on). I perform this step several times to create a few layers that I can play off one another using blend modes.
Then using Laminar’s blend modes, I adjust the percentages of each version that I’ve created until I have the right tonality and contrast level that I want. One layer may be set at Multiply, the next at Darken Only, and a third at Soft Light.
Once I’m happy, I output the image. If it’s rather small (many of mine are, since they represent crops from a full size image) I will bring it into iResize to increase the size to around 4,000 pixels square.
The resulting image is shown below – a mood created that did not exist in the original scene:
Note that iDesign and Laminar are only available on the iPad, while Photogene2, iColorama (as iColorama S) and iResize are available on both.
August 14, 2012
My approach to the conception and execution of iPhone art
My recommendation to everyone interested in improving their game is to read a lot of tutorials (on this site and other sites) and particularly to study some of the more advanced Photoshop books to understand layers, blend modes, masks and transparency. My "look" and "style", to the degree that I have one, depends almost entirely on heavy use of layers, blends and masks. Newer programs like Photoshop Touch and Laminar support multiple layers active concurrently - which is where the real power of creating highly textured, effective post-processed images lies. You can see my recent comparison of the two programs (from a layer-monger's perspective) here.
Studying art you love (for themes and the justaposition of elements) and photography (for composition) can be really helpful and can suggest a style, approach or subject manner to emulate.
And of course, experiment - repeatedly. I may spend hours on one image, doing 20 or 30 versions of it. Even then I may completely discard it and never return to it. I have at least a hundred pictures like this in my camera roll, and no one will ever see them. Ansel Adams famously said that if he captured one great shot a year, he'd be happy. He took a huge number of pictures to arrive at that one exceptional picture a year. Of course, Adams ended his career with well above that number per year. Still, there were a number of his pieces which didn’t see the light of day in his lifetime because they didn’t quite meet his standards, even if they met someone else’s.
For the image above, I've had the picture of the woman's face (output from Pixlromatic) on my camera roll for four months. I did at least forty versions of the woman alone, both as a single complete portrait and divided in various ways. I tried dozens of textures and effects. The second shot - of the startled man with a woman, was an image I shot just last weekend in a crowded Starbucks - and was about to delete because it was blurred and did not have a clear subject. Then I hit on the idea of combining his startled expression with her very intense look to create a "j'accuse" effect by the opposing of the two views, with her tight-lipped face above and piercing eyes below his startled look. I made his image even more blurred than it was and duplicated the less blurred version over the first and scaled it down. Then I photographed a rust hole texture and placed it as a sort of bullet hole in the forehead of his second blurred face - as though she could kill him with a look. Then, I went in search of a title / quote that would reinforce the message. Here's what I settled on for a title: "Quoth Shakespeare: 'And if you wrong us shall we not revenge?'", and this for the subtext: "No more tears now; I will think only upon revenge." - Mary Queen of Scots
There's a video by Dewitt Jones, a former National Geographic photographer ("Celebrate what's right with the world") that I have found very inspirational. It really expresses my view of life and my attempt to achieve creative excellence - it may be a bit cheesy in spots, but I think it's worth a view - if nothing else, to get at his idea of continuing to persevere, even after you already have that "good" picture you were looking for. When it comes to image processing, I look at the challenge of "getting to great" in the same way.
July 23, 2012
Combining images and illustrations to add mystery to your work
I can't draw a straight line to save my life, but I love to use type and illustrational elements in my images.
So, I haunt old bookstores, garage sales and swap meets for ephemera (old scraps of paper, pages from books that people consider trash, etc.) that I can photograph and integrate into my images.
During one such trip, I was attracted to part of a line drawing illustrating some author's belief in physiognomy as a way to determine an individual's personality. I thought it would make a great theme for an image. Once I had my "theme", it was a matter of taking a picture of a subject from the correct angle to allow me to use it, and finding a quote that provided context.
Here's the illustration before cropping down to the small area that interested me:
About a week later, I was having my usual Saturday morning coffee at Starbucks when I saw an interesting face that I thought would work well. I took perhaps 10 pictures of the man looking various different directions, and chose the one that most closely matched the orientation of the drawing.
Here's the image I selected, before cropping:
Then, I used Photogene2 to crop the image down to the ideal area, adjust the levels and to sharpen the image somewhat. It was very small and low res (about 300 x 300 pixels) so I knew I would need to apply a lot of textures to hide the pixelation. I then used Pixlromatic to get the right overall color cast I was looking for, and to apply a light vignette to darken the background. Then I used Photocopier to get it looking a bit like an old photograph. Whatever the photographer style I use, I always dial down the texture and grain and increase the detail matching so that the image is only minimally degraded during the color processing:
Here are the two final images - I finally decided on the one on the right as the starting point for the next step:
I then brought it into Photoshop Touch and upsized it to 2048 x 2048 (the maximum allowable image size in Touch). This is one of the few apps (other than iResize) that allows you to increase the resolution - and I'm very confident that Adobe's algorithm is the best one out there for doing so.
Using Laminar, I applied a series of photographed textures, as well as several internal textures. I always apply textures as a layer so I can later change opacity or blend mode as needed - or even duplicate the layer and apply a different blend mode to offset the earlier one. I then applied the illustration, cropped and adjusted to fit in Laminar. I set the blend mode to burn, at about 65% so that it would not be too dark.
I then used Quotationary for the iPad to locate a quote. I searched on keywords such as "physiognomy", "face" and "psyche" and identified one that works.
Here's the final result:
June 20, 2012
Creating polarized images with Xn View FX
Xn View FX is a straightforward photo editor with an unfortunate and easily forgettable name. Like several editors, it takes the approach of presenting you thumbnails that show you a particular effect, rather than expose the controls to create a cumulative effect yourself.
This can be a time saver if an application has the effect you’re looking for, and this one has a good variety. And the price can’t be beat – free. So worth a try. I’ve used it on three or four of my favorite images to good effect - here are two:
Here are some of features available:
* Crop, rotate, straighten, mirror
* Vintage, frame, texture, layer
* Gamma, brightness, contrast, hue, saturation, color balance
* Color effect, greyscale, sepia, ...
* Image effect such as sketch, charcoal, halftone, pointillize, crystallize, ...
* Blur, motion blur, tilt shift, glow
* Deformation such as symmetry, kaleidoscope, marble, waves
* Maximum, minimum, edge, emboss filters
For today, I want to concentrate on creating a posterized image, and then adding some light leaks and texture (halftone dots). You see which of the two “final” versions you like best.
First, I used the straightforward crop tool to size the image down to focus on the train. As you can see below, effects fall into four general categories that become pretty intuitive after use: Image Color, Effect and Filter. Underneath each of these heads, there are anywhere from six to sixteen effects to choose from. Below those (such as Image / Vintage), are eight to twelve variations in that category. Very easy to navigate and see in advance what your effect will be.
In this instance, I traveled down the Color / Hue branch:
And adjusted the color until I had the effect I wanted. Most effects have some controls that you can adjust.
Then I selected Image / Light / Enchanted bokeh:
The result was pretty satisfactory with no further adjustment:
But I also decided that it would be fun to play with Halftones (Effects / Halftone) and then Effects / Band.
Which do of the two above versions do you prefer?
June 5, 2012
"The limits of human calculus"
For me, some images I shoot are instantly usable, and I know exactly what I want to do with them – what effects I want to apply, and what the final look will (should) be.
At other times, some images don’t seem to work by themselves, so I save them for later. At some point, I find an opportunity to use them to create a composite image – sometimes a collage, sometimes a seamlessly integrated piece.
This particular piece consists of four images – two shot at one event, and the other two shot at different places and times:
This collage makes extensive use of Superimpose to mask and blend a number of recently shot images and photographed textures. Leme Cam and Camera Bag were used to create lomo versions of existing images.
Photogene2, my favorite general go-to image editor (because of its similar feature set to Photoshop 7 from back in the day), was used to crop the images of the young man, older man, bird and nail.
The nail was captured with my new macro Ollioclip lens, which has superb depth of field - tilt-shift wasn't required to get the blur toward the end of the nail shank.
At a local garage sale, I found a number of pieces of ephemera like the math calculations in a European hand. They add a nice aura of mystery to the image.
There is a pretty good selection tool within Superimpose (hidden under the Masks button) that allows you to fairly accurately crop to your subject. This allowed me to separate the two people from their backgrounds. I also made use of the Exclusion blend mode – which I’ve never fully understood – because at a low opacity setting, it allowed me to add some artifacting to the image that works like a texture and a color change at the same time. That’s what gives the collage elements their rough hewn, "red noise" appearance.