Saturday, August 05, 2017

Making Softer more Intelligent Filters

I have a long (aka deep) interest in the application of AI (artificially intelligent) techniques going back to the 1970s, it even survived the so called “AI Winter” of the late 90’s and 2000’s. It no surprise that I have been interested in the fairly recent Deep Dream applications to recognise style and transfer it to photographs. I have even, although slowly, warmed to “filters” that perform the magic of image enhancement (a lot of the early auto-fix buttons where and still are only half baked)

Trying to keep abreast of developments (and there a lot and they are fast), I noticed a very interesting presentation by Michaƫl Gharbi, a MIT researcher and also a keen photographer, at this years SIGGRAPH conference.

With this or similar apps I think you can expect smartphone technology to rival current expensive SLRs within the next couple of phone generations. However I think an intelligent camera just got a lot closer and  Michael has also been working on that also.

Thursday, August 03, 2017

Seeing inside the frame

I’ve happened upon on a couple of simple tools to really help you see the important aspects of tonal balance and composition. This approach is specifically for the analysis your existing photos, but the insight gained from using them to prepare some composition sketches is very enlightening and of majour benefit for your future photography. There may be some one click filters that undertake similar methods, but it is the act of doing these steps yourself that will not only improve your observational skills but with make looking for tonal balance an automatic habit.

The first step is too find a software tool, a simple photo editor, that will allow you to do the basic editing like lightening, darkening, [cropping]. It should also let you make a black and white version and possibly posterized versions. It is important that it can display the photo full screen (or as close to full screen as possible). In this case I am using, which is a great little user supported editor. Once you understand what’s going on the process, I suggest you migrate to a non-destructive editor. I  played around in Luminar & Polarr , they are also easy to use and do the job required beautifully. Picasa, if you still have a copy, is also perfect and with the added advantage of scrolling through your collection and also rating the photos you like. Lightroom or AfterShot pro could even be used here, if you are already familiar with them (don’t get them just to do these steps, eperiment with something free to start).


Load up a photo you the are interested in analysing, I’m using the Day 5 photo from last week a Foggy Afternoon in Olinda. The next step is to make it black and white. I prefer to just remove all saturation, but any convert to black and white approach is fine. The next step is to have close your eyes and look for the shapes in just 5 tones. You can lighten or darken the image to get the balance better. Only use the contrast if you must. Controls that let you adjust the highlights or shadows independently are preferable to just altering overall contrast. Now display that image full screen. Only save this version if you want to and be careful to give it a different name if you are not using a non-destructive editor.

The second tool is a neat little sketching program, called mischief, which allows you to draw over the photo, as if on a light table. There are several expensive graphics programs that allow you do something similar to this via layers but they are complex to use and set up. Mischief by comparison is really intuitive to use,  infinitely scalable and is no more complex than putting a piece of tracing paper over your photo and grabbing a pencil.   So I strongly recommend you try these steps with it. Looking on the mischief web site you will see a free and a paid version, you only need the free version to follow my method.Notan test Mischief starts full screen and there are a few controls you need to become familiar with so I have already made an Anatomy of Mischief post back a bit in the blog. Next change the window background opacity (you can use the Ctrl+W keyboard shortcut or the toolbar). I find it helpful to make a little Notan legend of 4 greyscales tones somewhere on the edge. You can later use the eye dropper to pick these shades later as you work on your “tracing”. Using the marker brush (its the bottom brush in the free version brush palette) and make it fairly thick and 100% on the opacity slider. Now start colouring over all parts of you photo that have the same tone. This is a lot easier than you might think. If you paint over something and you can not really see where you painted you have got it right. If you can clearly see where you have made brush marks, its time to back off, erase those marks, or use the undo tool (CrtlZ)  Think of yourself as doodling in the shades. It doesn’t have to be perfect it is about training your hand eye coordination to almost do it in autopilot. Turning off and on the background transparency now and again lets you check your progress. You should avoid getting too detail but you need to cover the main tonal regions (and arrive at something like the sketch on the right.

Now is the time for the most interesting steps. First look for the main positive and/or negative shapes (pick a colour to outline them and outline simple geometric shapes).image In this case the subjects (the positive shapes, my dotted light blue outlines) are the dark shadowy trees and the negative space is the central column of light. The four dark positive shapes are a little asymmetrically balanced either side of that. This is a great start.

Next look for the high contrast edges (and lines). A viewers eyes will involuntarily flow along these high contrast paths paths, because that's what our vision does and our brains use this information to make sense of the image. The direction of the eye will usually be from less contrast towards more contrast along these edges/lines. Also the eye & brain love to play the lost and found found game (lost and found lines and edges is one of a painters secret tools). It is best if these paths can converge on a “sweat spot” or lead onto another path. It is very important that they don’t lead the viewer off the edge of the photo, particular at the corners of the image. If that happens you risk the viewer moving on, scrolling down or turning the page.

Study your tonal doodles and compositional squiggles with squinted eyes. Then close your eye for about 30 seconds and half open them again but follow a different path or focus on an alternate center of interest. Close and half open your eyes a few times until you are following the same paths to the same sweat spots.

Are these what you wanted to show when you took your photo?

You really don’t have to save any of this activity. Its more important that you do it, Doing it as if sketching develops your skill to accurately observe tone and understand composition at an innate level as opposed to following sets of rules. You will be starting to see with an artists eye.

If you have been using a non destructive editor, bring up the colour saturation again (close to the normal mid slider setting) leaving your tonal adjustments. You will probably be amazed, the photo is likely to look less washed out and the colours look more natural (closer to what you remembered when you took the photo).

Wednesday, August 02, 2017

Time for a Software & Social Media Diet

Alvin Lfe coachMoving most of my time and tasks, much of which is now Art & Photography oriented, to reorganised workspaces and computers was more tedious than I had expected. I had managed to accumulate a lot software applications and tools which looked interesting at the time but I haven’t used since. I have to admit I also have a few social media accounts that are collecting dust and/or pretty much ignored.

Rather than wait till new year and a big resolution setting time I figured now is the perfect mpment to go on a diet, stop feeding the empty social spaces and slim down the bloat inside my computers. I’ve enlisted Alvin as my life coach.

I sure hope he knows what he is doing?

Monday, July 31, 2017

Getting up to some Mischief

My disdain for workshop or trainers that just teach their “work flow” as the ultimate (perhaps only) way to use software and carry out various photographic or artistic tasks is continuing to grow. Such attitudes are a guarantee to remove originality, innovation and individual expression. Those that religiously follow the “work flow” rules end up producing much the same work over and over again and whilst it can be good, even excellent it is soon an avalanche of sameness, blunting our interest. If you don’t believe take a look at someone scanning their social media. they thumbs or fingers are barely allow more that 2 second to view any image. Swipe. swipe, swipe….

I favour a lightly slow more honest evaluation of the tools, workspace and characteristics of the material. You still need some instruction in the basics but ability comes more quickly if you just play with the media and tools. Its not unlike many of the early craft guilds (which included painters, or newspapers in the old days when they had staff photographers). Here you where probably apprenticed to a master and they expected you to grind the pigments, mix the inks, wash the brushes and even sweep out the work shop. You got to handle all the media and saw how it was used and eventually you where guided into doing it yourself. Some apprentices adhered to the methods of their master but fortunately for us enough students experiments and expanded on what they had learnt and their masters where wise enough to encourage them. I believe the same approach needs to apply in using the modern tools or art and photography. The masters, can be those guru photographers and the You Tube Stars, you no longer need to be in their workshop but the familiarity with the workspace and the tools is still down to us, the “modern apprentices”.

I’ve told you all this because I want to resume short blog posts about the work space in software and the key tools to use, so that you can understand the steps in the work flows you are taught but also have enough confidence to experiment and try something of your own. I happened upon the Mischief sketching software whilst looking at ways to prepare some composition/notan sketches. It was a it was a real relaxation, simple to use and powerful at the same time, no work flow required. There is a free version which is perfect to get started with, and is what I am using here, If you an make a mark with a pencil on paper you good to go.

Mischief Workspace

The Mischief work space is a single screen and it normally starts in full screen mode. There is the normal windows style menus bar with drop down menu items that run all the programs function. Mischief Tool BarThe more commonly used are also shown in a vertical tool bar (normally over on the right). One very unique aspect of this work surface is the “canvas” you work on is that it is truly infinitely. You can scroll off the edge of the screen or zoom out to reveal more canvas. This an incredibly liberating feature. Start by drawing something that runs off the page, pan across a bit and keep drawing. The last two icons are for the the undo (another feature without a limit) and redo. All the tools and functions have keyboard short cuts, if you prefer those (look under help).

The next important set of tools are found in the brush panel (normally down on the lower left). The free version only has three brushes (aka pens) and Mischief Brushes Panelthey only have simple controls on size and opacity but this has been cleverly used to create a faint lines (pencil) for rough sketching a solid line (ink pen) and a thicker chisel style semi transparent line (Marker pen) These three pen will give you a decent sketching experience. probably the thing you will want to change most is the pen size, which uses the [ and ] brackets (for thinner & thinker) r the little slider pull out by clicking on the second bottom tab. the opacity can also be controlled by the number keys (again see the shortcut keys under the help menu.

The final on screen panel is the colour picker, normally on the upper left. Compared with many other packages this is a very simple format, but it does allow select for Colour (the slider under the colour square, Tone, the vertical position and Intensity, Mischief Colour Pickerthe horizontal position on the colour square. Unlike the paid version, the free version doesn’t have the ability to save your own swatch of colour, but size the canvas can be of infinite size you can place a small swtach of colours directly on just visible edge of the canvas, then return to that with the eyedropper to pick one of the few colours you have selected. So for simple composition sketch with 5 grey tone I have just used an approx 25% grey 50% grey, 75% grey & Black, using the white of the canvas as my 5th tone. You can either erase this later or just pan so that it is not displayed on the screen when you export a picture.

A really unique feature of this program (I haven’t seen another implement such a simple idea) is the ability to change the canvas opacity. This basically changes the canvas into tracing paper. Mischief Tool BarBy putting a suitable photo full screen and loading mischief over the top of it and either using the Window> Activate Window Opacity function from the menu or just use the slider on the side of the vertical tool bar. Moving the slider closer to 0% makes the canvas more translucent. You can trace lines of colour in the basic tonal shapes.

You really either need a Wacom style tablet of a touch screen with a pen, to get the natural feel of drawing. I have used it with the touch screen on my HP spectre and the touch mode on my old Bamboo tablet, but It is definitely wasn’t as good and trying to use the mouse is a little too fiddly. If you haven’t ever draw3 on a computer screen before the free version of the program is worth a trial. I’m already moving onto the Pro version (mainly to get layers & pins, more about those later hopefully. In the meantime never be afraid to experiment with your own “working methods” and occasionally just have a good “play”.

Saturday, July 29, 2017

Deeper into Disappointment with Google

Over the years google has bought onto the internet a number of great software tools. I have and still use many of them, Blogger, picasa, gmail, My Tracks, Google Earth, Maps, Google Drive (Google Docs, Slides & Sheets), Nik Software, Snapseed). I’ve gone right off google photos (but still put up with it to publish photos into this blog). A number of these have been depricated and whilst they still work google no longer offers downloads, or alternative software and/or no support. The one thing they have been good with to date is giving warning and letting you download any personal data.


I have taken I liking to the idea of exploring that place between art and a photograph, in particular what I have been Hashtagging #AIart (on Instagram). Google created the Deep Dream generator originally as a way to explain how their image analysis technique worked. It is based on training neural networks (a type of artificially intelligent data miners). As well as recognizing objects, animals, people & things the networks can also be used to seek out style (ok in a rudimentary way, it can mimic marks making, line works, colour palette and several other features that make artworks original. The software does not create original art but it can take a photo and find that style of linework & colour etc within a photos. Some of the result are amazing. Unfortunately more and more examples that Style formating selection optionsI see produced by other  “copy cat” system are degenerating into the garish and boring, as they become very popular of social media.

Clearly google are looking to muscle in of the social media aspect and they have add a multilevel aspect to site membership (its free) and that is tied to number of public post and more importantly number of likes. A different levels you can only generate a specific number of a specific type/style of image with a set time frame.  So It only took me a couple of images to reach my limit with a polite come back and try again later. The type of thing that just plain kills exploration and creativity. The writing is on the wall google this is going to become a ghost town like google+

The one nice thing, is the UI (user interface has been smarted up a bit with better layout of selection buttons and other tools. Nothing new just a little bit cleaner to look at.

? info panel within Deep Dream One new feature, that I have asked for a few times, is better documentation of what was set up  to produce each result. It seems quiet common that most people (especially those new to deep dream) blanket bomb one image with many alternative networks/styles, and after doing several and waiting a decent time for each it is difficult to remember how you created a particular masterpiece. The ? Intro item now gives you a run down although it has only remembered a few of my style images.

Perhaps its another case of needing more likes?

No matter, such is life.

So Why am I complaining, the software tools are free?deep dream of a Perigrin Falcon