Showing posts with label 4d. Show all posts
Showing posts with label 4d. Show all posts

Thursday, December 22, 2016

Painterly…

Winter is getting colder, that sort of thing I would call good news! Meanwhile, I'm trying to get involved more into my thoughts and myself, and my experiments; as it seems the only thing I'm good at currently, with the lack of go-outs to snap some pictures, or even working indoors. I guess I can say I'm investing more time into my thoughts rather than my work with the camera. In hope these thoughts will deliver me somewhere.

Escherism

Well, I've been interested in the art of this man long time before I even adapt myself to photography. I've been reading about him recently to see if I can get any clue to get me further with my experiments with the camera, and I was somewhat happy to see that he got his biggest spark of inspiration from the tessellations of the Moorish architecture in Spain. The aspect of tessellations is something I'm considering, specially after some experiments with some of my panoramas.

Which way did they go?

Probably my first successful trial with this trend of surrealism (which I like to call Escherism) was with a panorama from Aughnanure Castle, specifically from inside the small tower; a panorama done in my 2014 visit. Since a panorama is a merge of several images, what about merging several panoramas? Does that it make it a hyperpanorama? Anyway, I had this idea for a long time and my aim was actually to expand on the edges of a regular spherical (equirectangular) panorama, because these edges (zenith and nadir specifically) are usually compressed and so distorted. Thus, my trial was to expand on these edges and fix this distortion to have a normal view WITH a equirectangular panorama of 360x180 view.

Surreal Jabreen


In the beginning, merging was hard to do, since PTGui nor Photoshop were useful in this kind of process. Thus, things had to be done manually. Since it would be a manual work, I had to program myself and my workflow, and in return I've generated a workflow in forming these panorama of various projections. So far, only the spherical (equirectangular) projection is adequate for such a process, but I'm trying to experiment more. I've done experiments already in the field of Mercator projection, which was a lengthy process and a hard task while the end result was quite a shock! But well, some people liked it after all (on instagram); seems surrealism got an audience far more than I thought it would.

Let's talk Dadaism.
Using Mercator projection. The whole scene rendered incomprehensible it seems.


On the list of projection styles is, of course, the planet projection and maybe the circular. Not expecting much out of those, since the distortion is high, but it might be a plausible surreal work. Doing a single surreal work like this would require stitching 7 to 25 panoramas! Thus, my work so far is on a small scale (not exceeding 2000 pixels wide under best conditions), and all are in JPEG format; nothing in HDR like I usually like to do with my regular panoramas. However, for some technical issues, I might have to stitch small HDR slides and tone-map all the set with the same settings before working on merging them in Photoshop.

Methodology

I'll explain here the method (or some of it) which I've used to generate the images above. As I've stated above, I needed from 7 up to 25 panoramas, but I believe 7 is quite enough. To organize my thoughts on the order of the images, I've assigned coordinated points to each panorama and literally name each to its position (using the Cartesian system, i.e. X and Y) in relation to the center (0,0) point, and typically this point is given to the zenith (i.e. the panorama is centered around the zenith point, with 90 as a pitch value). Of course this is just my reference point; any point of view of the panorama could be used as a (0,0) point.

Imaginary points for organizing and setting order to stitched
panorama of different pitches and yaws.


With assigning that point and forward, the yaw value defines the "X" value, while the pitch defines the "Y" value. Of course these values are in degrees, and I thought that the value of 45° is just suitable for the rotations in pitch direction (i.e. vertical or "Y" values). This means, each 45° is a single unit. For example: starting from the (0,0) view (and after stitching the panorama at this view point), raising the pitch to 45°, that means arriving at (0,1), while raising it for another 45°, or going directly from pitch "0" to "90°" right away, thus the point would be (0,2); while doing in the negative direction, like -45° means going down to (0,-1) and so on. Same concept is applied to the yaw value but with the yaw it is the "X" axis which changes: going 45° in yaw would be (1,0), while -45° in yaw is (-1,0).
It is important to track the process and always turn back to the (0,0) point after few changes in yaw or pitch. I've made many wrong stitches at the wrong angles. This is one reason that I prefer to assign the (0,0) or center point to the zenith view since it is easier to recognize. Also, I started saving the stitched panorama with names reflecting their positions on the grid accordingly. Working with few panoramas right now, it seems that not all points are really required; mostly 15 stitched panoramas would be needed, from (X:-2,0,2) and (Y:-2,-1,0,1,2) mostly. Working with other projections, however, would require more, like in Let's Talk Dadaism above, which was rendered with Mercator projection, yet the scene is not pretty clear because of the distortion. Some people liked it that way though!
I've stated before that I blend these panoramas manually, but in fact it's semi-manual process, where I have to place the slides manually and overlap them over each other at specific points, and then use Photoshop's Auto-Blend Layer... command with Panorama and Seamless options on. Thus, the blending process is sort of random but turns fine with panoramas being in spherical projection (since distortion is far less than in other styles).

Kubismus

Been awhile now for me being interested in some of the artistic movements of the past (and which still exist and followed I presume). One of these movements is Cubism without doubt. It is considered one of the cornerstones of arts in the 20th century. I loved the idea and wanted to investigate how that is applicable (if it is) to photography. To my surprise, I've figured that one of the ideas I had in mind (but didn't achieve yet) was actually one form of cubism, namely shooting several shots of the same object at slightly different angles or distances and making a collage of them; some photographers did that already even back on film days! Well, maybe not new idea, but I was still going to try it when I get the time (and subject) to.

Cubic Selfie
Up until finding the good subject and the time, I've decided to edit and retouch some photos for a trial into cubism. I guess my favorite target was my selfie from 2015 (an infrared shot with a regular camera). The process was simple: open the image, copy the image into another window (or file) and make it small, then copy specific details from the large image onto the small image, with some addition of some details later (like shadows). Some people stated that it looks 3D-ish, and this is an important remark, because sometimes, all you need is just add a shadow to your work and you are at the steps of a 3D image!
Generally speaking, cubism is supposed to be about "neglecting" the perspective and considerations for light and shadow and simply capturing the essence of the subject, thus adding a shadow here does not "go under" the umbrella of cubism I suppose! But maybe things are different when it comes to photography and the way that a camera work; hope that point would plead to forgive my shadow addition when editing this.

Cubic Corrib

Things evolved a bit later, specially after discovering a set of photos from my 2009 visit to Ireland (my first) where I've shot several photos at different zooming levels for the lake and the boat. Thus, I've started taking details from the other images and paste them onto the first photo (with least zoom level). In addition, I've started adding more effects to the squares (cubes) like bevel effect, trying to approach some good level of a cube-like shape. Significant details here were scarce somehow, thus I had to tear down the basic image further into few squares to fill empty spaces. To me, Cubic Corrib is still not as much interesting as Cubic Selfie despite the effects put into it.


Cubic Pjazza

It is natural of course to think of panoramas and the cubic art, but instead of following the usual steps with the other regular shots, I've sliced panoramas and fit them into a cube (while making front layers transparent). The Cubic Pjazza cube was just an experiment. The main problem is not being able to show the inside or the back of the cube and lowering the opacity (by 50%) of the front layers (mainly front, right, and top layers) seems not quite a plausible solution since much details are somewhat lost. The next step was, of course, a hypercube!

Kylemore Tesseract

Again, we have a problem of opacity and unclear details, but further more there is a significant problem of filling the space between the two cubes (which are supposed to be planes connecting the two cubes, hence raising the dimension level). In Kylemore Tesseract not all the space is filled, but only the vertical planes connecting the corners of the inner and outer cubes. To add, I'm not sure how (or rather with what) to connect these 2 cubes: what kind of slides I should use? This thing is still in experimenting phase, but I do really think of making the experiment physical somehow by printing or creating models out of these panoramas (in slides), and try to put them together.
A funny coincidence occurred as I was working with these hypercube experiments, which seems to me that I've created my first Impossible Object!

Optika

It occurred to me as I was assembling a cube made out a panorama shot in Sultan Qaboos Grand Mosque in Muscat, Oman, in October last year. The bluish tint in Optika is generally caused by a shift in the color space but I kept it without fixing. Creating this by coincidence makes me consider photographic opportunities in the future. Not only that, but also considering other shapes and trends, like triangles and their properties, and, of course, circles as well.

Finale

It seems my brain is chasing its tail right now, specially with me dipping myself into math by the nose. Deep inside, the need for a vacation is increasing even further. All this indulgence in arts and math seems to be nothing but a trial to neglect myself and whatever tires my mind. In hope of creating something beautiful, though.
There are more ideas and even more sources of inspiration (from Sabah Fakhri as usual and even others - mainly Arabic songs and singers). I don't want to give excuses to myself, but having a quiet home and a good space to work in are now pretty much essential parts of having an easy flow of thoughts and work, yet, I'm lacking these two aspects which makes me reluctant and must think about every movement I do to move my camera gear here and there, beside the accessories. I guess I have to seriously start noting down any ideas that pass my mind, otherwise I will be at loss. Many of these ideas would require having a model as I can't (or rather it's advisable not to) be the model myself and the photographer in the same time. And of course, more props and accessories needed.
My next step in this life, so far, is to take each day as it is and suppressing, hard, my panic attacks and the fear for tomorrow. For in God I trust…


Stock photography by Taher AlShemaly at Alamy

Thursday, June 9, 2016

3+3=4…

I wasn't sure that I was going to type something in this blog for this week, but I'll try to pass this as fast as I can. Ramadhan is here already and my time is consumed greatly. By consumed, I don't necessarily mean "work" or anything else; but it could simply mean being tired and unable to do much.
After breaking the fast, I spend my time doing some usual chores and if I can get some time I would be working with my camera somewhat (if any idea was available). I'm observing this for the upcoming weekend, though I'm not optimistic about it with such family gatherings in weekends. I decided to fight sleeplessness and the upheaval in my sleeping pattern by reducing the amount of food I eat after breaking the fast, and I must take it at specific times. No fatty dishes, just dates and some bananas with milk (sour milk).

Meanwhile I was able to do some experimentation with some idea that I've been thinking about for a long time now and I'm finally on the move to do it. The idea is creating a so-called 4D image.
The problem here lies in the fact that I don't really have any imagination or any expectation for the outcome. I'm merely working by the concept and the theory of 4D space. That is, merging 3D shapes with each others would raise the dimension space further, to 4D.

4D Hexagon
an old experiment done by merging 2D hexagons with each other to form 3D Hexagon, then merging two 3D hexagons to form 4D hexagon, supposedly.
Probably this concept is best illustrated by the tesseract, or 4D-Cube. Anyhow, from this point and concept, comes the idea: what would happen if I could merge two 3D images? What I'm going to see, and what I'm expecting? Up till this moment, I don't really have any answer for this, and the main factor that keeps my mind busy is how to shoot, and how to merge?
As a principle, I've realized that I would need at least 4 shots. In the beginning I used my 100mm macro lens with a chess set, and using a shallow depth of field, but then I've realized that a shallow depth of field is not even a good thing to create a good anaglyph or 3D image. Thus, I've changed to 50mm and using f/8 (or around that). I've taken several lateral shots (sliding the camera to the side 1cm between each shot), and once again four other shots rolling the camera back to front. However, I ended up using the lateral shots alone. My idea was adding some depth somehow using back-to-front images, but it was puzzle that never worked out.

End Result

There were many trials and merging several layers by combining these 4 lateral shots. I won't go into details, but the major problem was how to merge or blend two 3D images? Not to say, upon which point (or focusing point) should I base the alignment of the two 3D images? I ended up using the front knight as you can see above, while blending the two layers by trial and error (ending with using the Darker blend).
For comparison, I've put the End Result above, which has a comparison between the regular 3D and the supposedly 4D, or the merge of two 3D. The two must be, supposedly, viewed using anaglyph glasses (red-cyan). I need some feed back on this, so if you can do this in the comments section I would appreciate it. I cannot rely on my own judgement alone on how this 4D looks or feels. As a beginning, it seems that the chess pieces were broadened and became thicker than the original, and as for the background it looks shaky. At some point in my experimentation before reaching this result, I've indeed made a version where the knight piece in the foreground was stable, while the other pieces looked shaky and blurred.

Finale

This is all I could put on for this week, and I'm not sure I will be posting for the rest of Ramadhan, but I would if I have any new ventures and experiments. Probably I should neglect photography a bit during this month and work more with my Geltani? Maybe.
Anyway, if you are reading this, and have the possibility to view the End Result above with red-cyan glasses, please leave your impressions in the comments. It would be a great help to understand how other people perceive this "trick"!
Time to manage my sleep now, again…