Luminous Landscape Forum
Equipment & Techniques => Cameras, Lenses and Shooting gear => Topic started by: teddillard on October 29, 2013, 07:17:42 am
-
Long, very long story short - I've finally got a really solid prototype of an easel that allows stepping capture by moving the artwork precisely, rather than moving the camera. I'd really be interested in hearing feedback on it. As far as I've been able to find, there's nothing else out there like it.
(http://xyeasel.files.wordpress.com/2012/03/img_1561.jpg?w=574)
In short, it's a vertical easel on the lines of a classic, large, H-frame painters easel. It's equipped with a stepping motor (so far, manually controlled, but easily automated) that allows precise vertical motion. It's on a dolly, which allows horizontal motion, also able to be motorized and automated at some point.
The short pitch is, by virtue of the camera remaining static, focused on a fixed frame, and the subject moving through that frame, that frame is the only area you need to light.
Here's a link to the site I tossed together to help show what it's all about: http://www.xy-easel.com. There's a video there explaining the basic idea there too - can't figure out how to embed it here. ::)
Looking forward to hearing your comments. If you'd like to PM me, feel free, or email me at ted (at) teddillard.com
-
I love this idea. Obviously, perfect lighting is paramount, and eliminating optical aberrations, too. I'll have to look into it.
A downside: It won't work when the art cannot be removed from its installation (which happens on occasion).
-
Long, very long story short - I've finally got a really solid prototype of an easel that allows stepping capture by moving the artwork precisely, rather than moving the camera. I'd really be interested in hearing feedback on it. As far as I've been able to find, there's nothing else out there like it.
Hi Ted,
It's looking good. It makes sense to use an inverted T-frame holding the artwork, sliding up and down on an H-frame. I agree that confining the area to be lit to a smaller area than the full artwork gives most flexibility and potential quality for lighting. It also allows better shielding of stray light, and of potential lens glare.
The only thing left is perpendicular alignment of the camera, but with flat-stitching that becomes slightly less critical since lens distortion can be corrected at the same time as optimal squaring. This will allow faster setup, although still no licence for sloppy work.
Well done.
Cheers,
Bart
-
Yes, I went through a few variations on holding the painting securely and finally settled on the same basic idea as what a traditional painters' easel uses. There's some lesson there about trying to re-invent the wheel, I'm just not quite sure what it is... ;D
On getting it square and parallel, that's not a problem. I just use a Zig-Align, and as long as the camera is at the same angle as the art, which figures to about 10º, you're perfectly square and centered.
As with any case of shooting flat art, a good lens really pays off. I was using an old 55mm Micro-Nikkor, but found a good deal on the 105 Micro-Nikkor and that's even better. Edge-to-edge sharpness a few stops down from wide-open is amazing.
-
If you haven't, you might take at look at Hugh's Easels, or Gung (apparently no longer available in the US). I painted for years on a Gung TJ7070B. Superb easel: x on rollers, y on a counter-balanced slide, tilt on a crank. I put the whole easel on three extra large, soft-rubber locking wheels (z). When I bought mine, the top-of-the-line supported something like 1,000 lbs. — made for working on sections of frescoed walls.
http://www.hugheseasels.com
http://www.artsmate.com/?product-98.html
-
There's also David Sorg's easel: http://studioeasel.com/ These are what got me looking at the traditional painter's method of holding the work securely. Unfortunately, a lot of those easels are very heavy, and very expensive.
I was originally planning on adding a motor drive control to one of those and making it a permanent, non-portable version along with a smaller portable version. As it turned out, the one I ended up with works great for large canvases, yet is still under 50lbs and is portable.
-
Thanks for the David Sorg link — good to have. :)
-
I use the same technique except the camera is mounted on a nine foot column. A counterweight makes it easy to slide the camera up and down. Of course, I shoot tethered. I've got a grid marked on the floor that's about 8 X 6 feet. Of course the camera is calibrated and a flat field lens is employed. The ceiling is painted flat black and the floor is dark grey. The pitch and yaw of the camera mount is adjustable. The 2 X 2 foot capture area is illuminated evenly.
-
What do you use to stitch your images? I find that all the stitching packages fall apart when there are large areas without much significant detail. I almost always revert to doing it manually in photoshop if this happens. (the best I've found is microsoft ICE)
-
Bob, the problem (and the key concept of the x-y easel) is that when you move the camera, you're moving the frame. The easel allows the frame to stay in place, thus, your lighting "stays in place" as well. You only light the frame you're shooting, rather than the entire work.
I've been using Photoshop to stitch, and yes, arranging most of it manually.
-
What do you use to stitch your images? I find that all the stitching packages fall apart when there are large areas without much significant detail. I almost always revert to doing it manually in photoshop if this happens. (the best I've found is microsoft ICE)
Hi Jeremy,
Dedicated pano-stitchers offer a solution for these types of image capture.
For example, PTGUI offers an Align to Grid (http://www.ptgui.com/examples/creating_gigapixel_panoramas_with_a_robotic_panohead.html) project setting, where the images that were shot in a particular order are placed in a grid formation. Subsequent alignment will optimize the tiles for which control points can be found, and will keep the featureless tiles in their shooting order position.
Cheers,
Bart
-
I should add - my next project is to use an Arduino control to program the stepping and shooting. If I can get it done (which really isn't that hard, Arduino-wise) then the entire system can be run literally at the push of a button - shoot, step, shoot, etc, all with one activation, and at whatever stepping intervals you want.
This is the reason the motor drive is such a key to the whole thing, and another reason why the big studio easels weren't working for me.
-
Bob, the problem (and the key concept of the x-y easel) is that when you move the camera, you're moving the frame. The easel allows the frame to stay in place, thus, your lighting "stays in place" as well. You only light the frame you're shooting, rather than the entire work.
I've been using Photoshop to stitch, and yes, arranging most of it manually.
I move the art, not the camera. The column simply enables me to move the camera up or down depending on the size of the original art work--anything from 24mm X 36mm to 6 X 8 feet.
-
Ah, got it, sorry. How do you hold and move it?
-
Ah, got it, sorry. How do you hold and move it?
An array of aluminum jigs with tick marks along the x-axis is attached to the floor with either weights, velcro, or suctions cups depending on the size. I also have a sheet of masonite that is designed to be used as a pin registration stage.
Along the Y-axis, another set of aluminum jigs are held in place by inserting pins into socket holes (hollow .75" metal cylinders) that have been carefully installed into the concrete floor.
The system took a few days to design and about a 40 hours to build. Here is an example of an artwork reproduced using my setup. http://www.topdogimaging.net/about/fine-art-reproduction.html I am now out of the art repro business and am concentrating on resuscitating antique and vintage photos.
My solution is by no means portable. It is accurate, simple, and precise.
-
Ah, nice. I have a few clients with huge murals, and one of my motivations was to make it so I could "come to the mountain" rather than ask my clients to bring the mountain to me... ;D
...I also very much prefer shooting in their studios. But that's another tale.
-
Well, push has come to shove, and I've put a Kickstarter project together as a "proof of the pudding". I've opted for this for a few reasons.
First, I firmly believe this is a really important development in the tools we have available for Fine Art repro - not only the easel, but the whole process. I want to get the story out there, and I want to do it with a complete body of work. As important, I need to do it with an artist who understands what we're trying to do, and is as critical as we are, and understanding of the technical aspects of the process, to really take this to the next level.
Warren Prosperi has been with me every step of the way with this process, even to the point of helping me develop it. Thus, The Prosperi Studio.
I also want to use this opportunity to help Prosperi Studio by reproducing their complete catalog of work. It's important work, and I feel it deserves support.
Read more about what we're doing here: http://www.kickstarter.com/projects/952491968/the-atelier-print-the-prosperi-project
I'd love some feedback and discussion about this. It's always been a work in progress, and even more now. This is a combination of the easel and stepping/stitching techniques, color management from front to back, and most importantly, lighting techniques. Of course, if you'd like to support the KS project with a donation or by spreading the word, I wouldn't mind that either. :D
-
HA! Just got this message: "Warning: this topic has not been posted in for at least 120 days.
Unless you're sure you want to reply, please consider starting a new topic."
No kiddin'. :D (I don't know if I'm just slow, or what, but I've been chipping away at this for over 5 years now.)
Well, I've been shooting with this thing and working on the mechanism and shooting and cursing and checking Craigslist... and this is the result.
(http://xyeasel.files.wordpress.com/2012/03/img_2491.jpg)
After poking away at this thing I found a few models of artists' easels that have an assembly that holds the painting and moves up and down in one piece. Seems obvious, right? Unfortunately, the classic artists' H easel isn't really set up like that, and the ones I found that were, were enormous and expensive. There are a few varieties, some can be modified and some come that way out of the box. This easel I found on CL and needed some modification. Even with all the stuff on it, it's under 40lbs and folds to about 6" flat.
I've found a couple of other things. Raising and lowering the work really needs some sort of lifting mechanism. It's just not safe to have a big, valuable painting on the thing and try to muscle it up or down, and it's very difficult to do so at any sort of precise increment. If you're using a live-view system, you're usually standing between the camera and the work, as well. The motor drive solves this beautifully, but it's expensive and heavy, and you need to plug it in. This one uses a hand winch from an ATV trailer - about a $30 part, and does a fantastic job.
(http://xyeasel.files.wordpress.com/2012/03/img_2492.jpg?w=574)
The wheel assembly is just a set of rollerblade wheels. They work amazingly well. I've mounted them with a good bit of friction so the whole thing doesn't roll by itself, or move when I'm cranking. I have one more addition to the thing - a clamp that will lock the wheels.
One seemingly minor addition, but far from unimportant, is a safety clip and wire to keep the work from falling off if somehow it jars loose. And I've had that happen.
(http://xyeasel.files.wordpress.com/2012/03/img_2502.jpg?w=574)
I'm going to do a little more formal design on the wheel assembly, actually putting a little trolley together that can be easily mounted to any easel. I've found that about 90% of the work I shoot only needs about three feet of horizontal movement, so you scale that out with a simple tape measure or yardstick.
The motorized version hasn't changed too much, except the addition of a motorized horizontal drive with remote control. The truth is, that's nice, but all that stuff adds weight, and most of the time I'm working on location and I'm not getting younger. ::) I'm not too far from pulling in some robotic guys and getting help programming and automating the whole affair, but to tell the honest truth, as cool as that would be, it makes something that's fairly cheap, light and elegant into something expensive, heavy and complicated, and for what? Just the part about not having to double-check for all my cords and remotes and all that, when I'm packing up is enough to put me off that direction.
I have lots of photos of both versions on the site: http://xy-easel.com/
I'm actually thinking of putting a how-to together, or selling kits, or even selling assembled easels. I could make a little drinkin' money on the side if there's interest, but anybody handy with a drill and a screwdriver with a sharp eye could probably put this together...
-
Ted, I have some large wall mounted pieces, but they would be a total pain to take off the walls along with their lighting brackets to take somewhere for someone to shoot. I'd probably bust the expensive museum glass in the process too unless I came up with a rigid crate of some sort for shipping. Plus, shipping them back and forth would be a fortune too like some find with large canvas-wrapped prints that exceed the UPS box dimensions and then need to go freight.
What if you did the reverse like in cinema? Put the camera and lights on the dolly and track and moved them left/right and up/down verses moving the art easel? Sort of an X-Y axis copy stand. That way you could go to the place to shoot it, and not someone having to take their stuff off the walls or frames to take somewhere and maybe damage them in the process.
SG
-
Obviously you have to weigh the practicality of the various solutions, but here's the thing. When you move the camera, you have to light the entire area of the artwork. I just don't see any practical way to move the lights in sync with the camera, and when you consider ambient light in the room, it becomes impossible in my opinion. You could do it with a room-sized dolly I suppose, but then you're back to a practicality and mobility issue.
For example - when I shoot a painting I typically use a single gallery-style halogen downspot in conjunction with "North Light" style daylight-temperature window/skylight fill. Because it gives me a result very similar to the artist's studio and most gallery displays. My light is at about a 8" height. I simply don't have the vertical room to move that another 4' higher - not many places would.
It is, if you think about it, the basic idea behind how a flatbed scanner works. Keep the lights and the lens in one relative position, and move them in respect to the work as one unit. Great idea for a small flatbed, but the bigger you get, the more the technical challenges multiply, and not really a great solution for a room-sized "scan".
-
Ted:
I have a Foba camera stand. Works great for up & down movements. The cross arm also has good range, but you have to make sure it's parallel to the wall. I like to have my lights at least 6' away from the art. Move them up and down to sync with camera if it's a large piece of art.
-
I wonder about using some LED light panel maybe with a shoot-through hole in the middle (Like a beauty dish.) and moving it in conjunction with the camera on a rail (or pipe track laid out on the on the floor)? Maybe two panels with LED wings on both sides of the lens?
I just did a UPS freight calculation on shipping a canvas-wrap of 10 feet long by 3 feet tall wall mount (40 pounds) and it ran about $490 one-way with UPS freight, then another $490 to get it back. Don't know, but $1,000 for a photo of it seems a lot even without the photographer's charge. I'd rather them to shoot it on the wall than deal with shipping the thing. Would probably be less too and less damage may occur that way. Trying to get it into a boxed/fitting crate is another matter.
Seems easier to have someone shoot it on location with some sort of camera and light moving gear. Maybe even a shift lens would suffice.
SG
-
SG:
Obviously you haven't done much copying. To have the light source so close to the lens, you are asking for a lot of reflection problems. Yes, you could couple the lights to the camera, but you should have them at least 45 degrees off lens axis, the further from the art, the less likely the reflection problems.
-
Well, John - there you went and put your finger on it, however maybe unintentionally. It's all about the lighting, and traditional "copy" lighting, though always what we were taught, is just not the best. You've got to consider the art as a 3D subject like anything else, and light it accordingly. Back in the days of film, we didn't have much choice but to shoot that way, and it was the only way to get even coverage of a big, mostly flat field, but digital photography allows us to take small samples, lit precisely, and apply that lighting evenly across the field of the subject.
And that's where the x-y easel comes in, along with stitching and field-evening software.
I've written a bunch about it, which you can read here - this is a good introduction with some samples: http://teddillard.wordpress.com/2013/04/22/lighting-in-the-artists-studio-fine-art-reproduction-as-photography/
But the whole mess is here: http://teddillard.wordpress.com/category/fine-art-reproduction/
Probably the best place to start is here: http://teddillard.wordpress.com/2011/10/24/on-fine-art-reproduction-or-youre-doing-it-wrong-part-1/
-
SG:
Obviously you haven't done much copying. To have the light source so close to the lens, you are asking for a lot of reflection problems. Yes, you could couple the lights to the camera, but you should have them at least 45 degrees off lens axis, the further from the art, the less likely the reflection problems.
Irrelevant argument as to Ted's idea of "Moving the art to the fixed camera," verses the opposite of "Moving the camera to the art." Who is going to submit "wall art" to a guy with a camera and pay all the freight costs to do so verses someone who can do it on location? Think of cinema: Actors can remain on the stage set while the camera boom moves around them, even if it has lights on it as well.
Having used old copy stands with 10" polarizers over the lamps and lens, it's not a problem moving the work under the fixed camera and lights on the table, be it lights on two or four on goose-necks. Exposure is already set. Simple snap-and-go. One could do it with bat-wing LED panels as well. Most pro camera shops already have that same setup.
However, having a customer take something large, and possibly costly off their wall, pay the high shipping costs and insurance and allowances for breakage, makes it a bigger problem than having someone shoot it on location and not bother the art at all. Think of shooting art in a gallery. I doubt if the Huntington Gallery in LA will ship you their "Blue Boy" painting, against them allowing you to shoot it on location.
Hollywood long ago figured it was easier to move the camera than the mountain.
SG
-
Just to clarify, the whole idea is based around shooting on location, not shipping the work to the photographer. The latest design is only 40lbs, and folds to 26x26x6". No tracks, no sets, no rigs, just the easel, a tripod and your lights. Because I'm shooting a small 16x20 frame, the lights at most amount to 2 or 3. I carry the whole thing in a compact sedan with a fold-down rear seat.
Take a look at my links regarding your lighting ideas. I just don't see it as the way to light the work.
As far as moving "the mountain", I can't figure that something mounted on a wall can be that difficult to take off the wall, except in some very unusual circumstances. Certainly not to the point of justifying building an entire movable set... for only one painting. ::)
-
Thats amazing! You actualy just build a design that's been in my head for the last year:-) :P
I do a lot of this sort of work and this does seem to address a very specific issue. I also thought about a wall mounted version that would keep the artwork parallel to the sensor plane.
Nice build!
John
-
Thanks John!
Funny thing, the interwebs... In the last two weeks I got two emails asking about this. Here's a little update.
I've refined the "Basic" design a little, with better wheels. The wheels mount on the sides, so they don't interfere with the folding, and one has a light brake on it to keep the thing from rolling unintentionally. I added some updated photos here: https://xyeasel.wordpress.com/2012/03/28/the-x-y-basic/
The "Servo" model still doesn't have programmable automatic controls, but it does have remote control of both axes now. I'm absolutely totally spoiled by that. I use it almost every day in the studio, and being able to position the easel and painting while peering through the camera? Just awesome. Right now it's a key-fob control, but I'm working on a joystick. Because, you know. Joystick.
There's video of that in use here: https://www.youtube.com/watch?v=7TKmXbjIO5M
The Basic is about 95% of where I want it to be to go into any sort of limited production, and I did end up selling the prototype to a photographer who was very eager to use it... This gives me a chance to get some feedback on the design, from someone other than myself. :D
Stay tuned, I'll update as we go.
-
Ted, I think the idea would be to light only the portion of the artwork covered by a single shot (well, there would be some spillover to ensure even coverage). This would be closer to a copy stand flavor of lighting, of course.
-
Um, sorry, I'm not getting your point. (Didn't I say that? Oh yes, on my OP: "The short pitch is, by virtue of the camera remaining static, focused on a fixed frame, and the subject moving through that frame, that frame is the only area you need to light." ;) )
Yes, of course you're only lighting the area covered by the shot, that's the basic idea that makes this such a powerful tool. ...but copy stand shooting? :o
-
Push came to shove and I just put in orders for parts for the v2.0 easels - both the Basic and now what I'm calling the Power Easel. Stay tuned for the update.
The Basic has a front crank for the up/down movement, with a dolly for side movements. I'm going to make the dolly available separately too, so you can use it with your existing easel if you want. I'm also going to have a remote-control powered version of the dolly.
The Power Easel will have remote side-to-side and rise-fall, controlled by one keyfob remote. It's going to take me a bit more time to fab this up, I'm hoping for a month or so before I have a working prototype... I'll update with progress.
Thanks for all the support and interest!
-
Push came to shove and I just put in orders for parts for the v2.0 easels - both the Basic and now what I'm calling the Power Easel. Stay tuned for the update.
The Basic has a front crank for the up/down movement, with a dolly for side movements. I'm going to make the dolly available separately too, so you can use it with your existing easel if you want. I'm also going to have a remote-control powered version of the dolly.
The Power Easel will have remote side-to-side and rise-fall, controlled by one keyfob remote. It's going to take me a bit more time to fab this up, I'm hoping for a month or so before I have a working prototype... I'll update with progress.
Thanks for all the support and interest!
Ted, thanks for keeping us up to date with the developments. I'm following the progress with interest.
Cheers,
Bart
-
Thank you sir! A little tease - here's the easel, without the drive mech: https://www.instagram.com/p/BDqAgr6tbYU/?taken-by=lifeinflux and the screw feed: https://www.instagram.com/p/BDqAngHNbYi/?taken-by=lifeinflux
Parts for the drive system are coming in from Hong Kong ;D
-
Progress! I have the dolly put together and testing in progress. Got some goofy vids up on the Instagram: https://www.instagram.com/p/BEBMUuGNbeJ/?taken-by=lifeinflux
-
OK, I have the Basic model together and going through testing. Here's what it looks like:
(https://xyeasel.files.wordpress.com/2012/03/img_3852_1.jpg?w=614&h=1045)
This is a HUGE improvement over the original design, using a classic front-crank with an acme screw feed for the vertical movement, and much stronger, more rigid support for the wheels. The guide track is also a lot slicker and simpler, and easily expandable, though not even as necessary with the new wheel configuration.
A lot of the drive upgrades come courtesy of the burgeoning robotics guys, and the direct-drive motor for the wheels, and geared motor for the screw feed is straight out of Battlebots. :D I've got the horizontal drive working like a champ, and the vertical screw drive just needs the coupling machined and it will go into testing.
(https://xyeasel.files.wordpress.com/2012/03/img_3857_1.jpg)
If anyone's interested, the Basic can be delivered in about two weeks, for $600 plus shipping. PM me, or email me at ted (at) teddillard (dotcom)
Read more here: https://xyeasel.wordpress.com/, and the accessories should be posted in a few days.
-
Got the radio control module working:
https://www.instagram.com/p/BE06ex3tbTa/?taken-by=lifeinflux
-
Got the radio control module working:
https://www.instagram.com/p/BE06ex3tbTa/?taken-by=lifeinflux
Hi Ted,
It's looking great. Is the wheel drive a single or multiple wheel construction. If multiple, which wheels (diagonal)? I could imagine that with enough friction between the wheels and the floor, the rails are of lesser importance, but with a single wheel drive the parallel motion is at risk.
Cheers,
Bart
-
Thanks again!
It's a small single wheel drive. It takes a much smaller, less powerful motor than I first used, which makes sense since it's pretty easy to roll around on those wheels. ...but for a piece that needs several captures it will crab-walk whatever you do - no matter how much care you take with aligning the wheels. The track is pretty much essential. Also it makes up for any unevenness in the floor, and allows you to use it on a carpet. I decided on a 4' basic section, with the ability to add sections and add a back track too. I can supply 10' sections too, but in the years I've used it I've found that large horizontal pieces are pretty rare.
About the only way to get a good synchronized more-than-one-wheel-drive is to use an axle, but then you have to run belts or chains, and all that mess. Matching two motors independently is really hard (edit: actually, impossible).
The only thing that's holding me back at the moment is the coupling for the vertical screw drive to the motor. There are about six ways to do it, just trying to decide the best.
I also think I'm going to opt for a bit faster horizontal travel. I erred on the side of slower, with my calculations, and I'm going to see how it handles a painting, but at this point it seems a little slow.
I'm also working on a battery option that will let you use a 12V DeWalt power tool battery instead of the plug-in power supply I'm using now. The cord is a pain in the butt.
-
I just figured out a solution to this problem that's been bugging me since day one. When you're setting up the camera and the painting, the best way to go about it is to drop the painting down to the lowest position and frame the camera on the top of the painting for the first row of captures. That keeps you from having to raise the camera if you do it in the reverse order and at the end, you've miscalculated and the camera can't reach the top of the work. (Clear as mud?)
This makes using any sort of a scale a problem, since paintings of different sizes with require different starting points as you raise the painting - the bottom rail starts in a different place. Solution? A sliding scale, using a simple tape measure and some of those super-disk-magnets. Check it out:
(https://xyeasel.files.wordpress.com/2012/03/img_1160-copy.jpg)
These are the magnets:
(https://xyeasel.files.wordpress.com/2012/03/img_1159-copy.jpg)
Line up the work in the camera, slide the tape to whatever mark you use to reference the position of the bottom bracket, and step in whatever increments you want - if you do 10" or 12", it's a slam-dunk.
-
Hey Ted:
Did you see what Google has come up with? The "Art Camera"!
https://www.youtube.com/watch?v=dOrJesw5ET8
-
Interesting... though just another high-resolution single capture device that requires lighting the entire work, am I right? 8)
-
Interesting... though just another high-resolution single capture device that requires lighting the entire work, am I right? 8)
Hi Ted,
I agree, that's what it looks like, just high resolution. And given the time it takes to get an exposure, I wouldn't be surprised if it was a scanning device, requiring constant illumination and avoidance of any vibration during the long exposure time.
The benefit of a moving easel is that the illumination of a relatively smaller area can be optimized for uniformity and color (and even use the best part of the camera's image circle, which also allows to reduce flatness of field issues), and resolution can be varied by using different focal lengths and distances.
Cheers,
Bart
-
Interesting... though just another high-resolution single capture device that requires lighting the entire work, am I right? 8)
Ted, I don't know. I like your set-up, and I think it's a good design for average sized pieces. I have an acquaintance at the Nelson, in KC, and am trying to get a hold of him to get some first hand information about the Google Art Camera system.
-
Hey Ted:
Did you see what Google has come up with? The "Art Camera"!
https://www.youtube.com/watch?v=dOrJesw5ET8
Funny. Taking a close look at the video, it's a great example of how not to light a painting. I'm guessing standard copy-board lighting, based on how the canvas reads. There's not accurate texture of the brush strokes, certainly nothing that feels like a North Light studio, or even gallery lighting:
(https://evmc2.files.wordpress.com/2016/05/screen-shot-2016-05-19-at-5-47-42-am.png)
The resolution is great, sure, but like the Cruse, resolution without correct lighting won't give you the feel of the work as the artist saw it. I wrote more about that here: https://teddillard.wordpress.com/2013/04/22/lighting-in-the-artists-studio-fine-art-reproduction-as-photography/
-
Right - read more about the Google project here:
https://www.google.com/culturalinstitute/about/
It's not about faithful reproduction technology, it's about what Google does - making information available to people. ;)
On the one hand, it's a great initiative, but on the other, it could be so much better. A friend of mine is the Director of Photography at a major museum here in the Northeast, and he laments the same thing - they want photos of the entire collection for their online catalogs, but don't want to pay/bother to do it right. :o
-
Right - read more about the Google project here:
https://www.google.com/culturalinstitute/about/
More info in DPR's yesterday's article:
http://www.dpreview.com/news/1947958268/google-art-camera-uses-robotic-system-to-take-gigapixel-photos-of-museum-paintings
-
I'll be really happy when "high-quality digitizations..." means high-quality photography, not just the highest possible resolution. Just sayin'. 8)
-
Updates - I've been sidetracked by another project, but while shooting the other day I realized I use several things almost every day - especially the X-Rite Passport and the magnetic easel, AKA white board. I just posted them up on the bottom of the Accessories page, here: https://xyeasel.wordpress.com/category/x-y-accessories/
Once I get a motorcycle out of my shop (that's another story... lol) I'll get back to work on the lift drive for the new easel. In the meantime I saw this on the way to work:
https://www.instagram.com/p/BGMs4pJNbXi/?taken-by=lifeinflux
(https://external-lga3-1.xx.fbcdn.net/safe_image.php?d=AQDVire_69fulC47&w=487&h=487&url=https%3A%2F%2Fscontent.cdninstagram.com%2Ft51.2885-15%2Fe35%2F13267336_977077859077899_371663974_n.jpg%3Fig_cache_key%3DMTI2NDU4MzAwMTg5OTM4MjI0Mg%253D%253D.2)
I asked the guy what it was, and he said a radio control for this:
https://www.instagram.com/p/BGMtCPmtbXy/?taken-by=lifeinflux
(https://external-lga3-1.xx.fbcdn.net/safe_image.php?d=AQBl6O3blkx-MDje&w=435&h=435&url=https%3A%2F%2Fscontent.cdninstagram.com%2Ft51.2885-15%2Fe35%2F13266818_478189455707707_683195715_n.jpg%3Fig_cache_key%3DMTI2NDU4MzY2MTY3MTc5MDA2Ng%253D%253D.2)
He can run the lift from across the street, where he can actually see the load. Welcome to the 21st Century! By the time I get this thing built, it'll be the 22nd! :o
-
Some more updates. I've been chatting with TTi, the vacuum easel guys, about some modifications for their big 30x40 model to work on the x-y Easel, and also with zig-align, the AMAzing tool for making sure your camera is square to the work. It makes a YUGE difference in how well it stitches. I'll be posting links to the site as soon as I have a chance.
In the meantime I got the 12V DeWalt battery adapter set up, and mounted the motor controller. Did another silly movie too:
https://www.instagram.com/p/BGhFqvctbWH/?taken-by=lifeinflux
One last step will hopefully be complete next week - coupling the vertical feed motor. Parts are coming, progress is being made! ::)
-
Ted, thanks again for the updates.
Some more updates. I've been chatting with TTi, the vacuum easel guys, about some modifications for their big 30x40 model to work on the x-y Easel, and also with zig-align, the AMAzing tool for making sure your camera is square to the work. It makes a YUGE difference in how well it stitches. I'll be posting links to the site as soon as I have a chance.
Aligning the easel parallel with the sensor with the Zig-Align tool is one thing, but how sure are we that the surface of the artwork is parallel as well? With an unframed sheet on a vacuum-board that would be reliable if we can vacuum the Zig-Align as well, but framed artwork is slightly more iffy.
Looking forward to what you come up with.
Cheers,
Bart
-
I use the zig-align on the art itself - set the painting up, set the mirror up hanging from a cord, resting square on the painting.
But still, considering the field of view is about 16x20" (using a D800, and going for the dimensions of the file at 300ppi), and the camera is at about 50-60" using a 105 Micro-Nikkor, your margin of error is negligible. Even if you were 1/4" off, one side to the other, (which you'd have to be drinking to be... :) ) that'd be what? .005% error?
Even when I'm lazy and just eyeball it, at f8 (about 2 1/2 or 3 stops from full wide open - optimum sharpness and even-field, from my testing of this particular lens) there's never any sharpness or focus issue, and the squareness after stitching the captures is always within the capabilities of Photoshop to correct (Edit>Transform>Distort)
-
I use the zig-align on the art itself - set the painting up, set the mirror up hanging from a cord, resting square on the painting.
But still, considering the field of view is about 16x20" (using a D800, and going for the dimensions of the file at 300ppi), and the camera is at about 50-60" using a 105 Micro-Nikkor, your margin of error is negligible. Even if you were 1/4" off, one side to the other, (which you'd have to be drinking to be... :) ) that'd be what? .005% error?
Even when I'm lazy and just eyeball it, at f8 (about 2 1/2 or 3 stops from full wide open - optimum sharpness and even-field, from my testing of this particular lens) there's never any sharpness or focus issue, and the squareness after stitching the captures is always within the capabilities of Photoshop to correct (Edit>Transform>Distort)
The Nikon D800 is your weakest link in the chain. I wrestled with it for six months--never could get it to render "accurate" color. The best 35mm option today is the Pentax K-1. The files are stunning and it bypasses bayer interpolation when it is set to sensor-shift mode. I've been getting phenomenal results using an Actus view camera, a Rodagon APO 80mm f/4 N enarger lens and the multi-shot mode on the Oly E-M5 II. The color accuracy and DR of the Oly are nearly comparable and in some ways better than the files from the Hassey CF39-MS/Copy stand setup I used from 2008-2012. The Actus/Oly/APO 80 lens combo comes in at round $3K. I use a Starrett 135a pocket level for truing up the art and the camera. I'm retired now and use this kit for personal projects and helping out friends document artwork and documents. The EL is a flat field lens with excellent geometry and color rendering characteristics. Better lens choices are the Schneider 72mm Digitar/Copal 0 (shutter stays open) or the Rodensock Sironar 70mm. I'm sure there are excellent shorter focal length lenses in the Scheider/Rodenstock lineups. https://www.amazon.com/Starrett-135A-Pocket-Nickel-Plated-Finish/dp/B0002FS0XM?ie=UTF8&*Version*=1&*entries*=0
-
The Nikon D800 is your weakest link in the chain. I wrestled with it for six months--never could get it to render "accurate" color.
I've never had a problem. I use the X-Rite Passport to set up a camera profile in ACR, and get pretty much spot-on predictable color every time. But then, my definition of "accurate color" depends on how fast and easily I can get a print that's faithful to the painting. ;)
-
Got both motors installed and coupled, though the vertical drive needs a little fussing:
https://www.instagram.com/p/BGzCly4tbTY/?taken-by=lifeinflux
I can't wait to get this thing into the studio! It beats the heck out of the old version... and I made a few adjustments that give me about 6" more vertical travel as well.
-
Got both motors installed and coupled, though the vertical drive needs a little fussing:
https://www.instagram.com/p/BGzCly4tbTY/?taken-by=lifeinflux
I can't wait to get this thing into the studio! It beats the heck out of the old version... and I made a few adjustments that give me about 6" more vertical travel as well.
Congrats. Looking cool ...
Cheers,
Bart
-
Got both motors installed and coupled, though the vertical drive needs a little fussing:
https://www.instagram.com/p/BGzCly4tbTY/?taken-by=lifeinflux
I can't wait to get this thing into the studio! It beats the heck out of the old version... and I made a few adjustments that give me about 6" more vertical travel as well.
Hi Ted,
I'm wondering how large a painting can your X-Y easel accommodate?
-
Hey Peter - the horizontal is pretty much limited only by your room and track length. I've shot paintings over 6' wide. On the vertical axis, the travel is about 40", but it can hold an 80" tall painting. I have to do the math, but I reckon the biggest painting would be about 60" tall without having to reposition the work. I routinely shoot 60" tall paintings with my working rig now, but a lot of it depends on the stepping overlap you prefer and the resolution you're going for.
Sorry, long day. Clear as mud? ;)
-
Happy Father's Day!
I did some math on the new rig, and I'm coming up with a 55" to 60" maximum height for the painting, if you're shooting with a 16" x 24" frame area. We'll see how that pans out in real life.
The original plan was to turn the painting 90º if I needed to for a taller vertical. I shoot a lot of portraits, and those guys like to go tall. In that case I'd just light it as if the side is the top, if you get my meaning. (I typically light to simulate a North Light studio with down spots, to emulate most painters studio lighting.) As I said, the horizontal size is really only limited to your room size, since it's rolling on the floor (with, or without the tracks).
The fact is, in actual use I've never had to do that. (I shoot artwork almost daily - I'm not retired, thanks BobDavid, but not for lack of tryin' ::)) The really huge work I've shot hasn't exceeded 60" tall, and is more often just really wide - contemporary abstract artists in particular, but also mural work.
The big work, though, is where you really start to get your head around the advantages of moving the work rather than the camera. Compare, for the moment, the difference between lighting a 5' x 10' mural, and a 16" x 24" frame area. amirite? ;)
-
Simple things make me happy. Like this:
https://www.instagram.com/p/BHXDt-MhvqE/?taken-by=lifeinflux
Believe it or not, it's taken me quite a while to put together a drive coupling for the feed screw that I liked. The bracket holding the motor is just a temporary setup, but the coupling works like a champ! (I'm sure anybody who's played with battlebots is laughing at me right now... :) )
-
Good luck but to me it is a solution for a problem that doesn't exist. Unless the artwork was huge I would always capture in one click.
-
Good luck but to me it is a solution for a problem that doesn't exist. Unless the artwork was huge I would always capture in one click.
Hi Bob,
Maybe you've overlooked the explanation that it produces higher quality images because a) one can stitch image tiles that were taken with higher magnification/resolution, and b) a smaller section can be better lit with constant angles of illumination.
Cheers,
Bart
-
Hi Bob,
Maybe you've overlooked the explanation that it produces higher quality images because a) one can stitch image tiles that were taken with higher magnification/resolution, and b) a smaller section can be better lit with constant angles of illumination.
Cheers,
Bart
No, I wasn't.
If you are doing art reproduction then you have a camera with the resolution to produce a print the size of what that easel is capable and you should be able to evenly light it.
To me stitching is something you doing when you really have to because it introduces distortion. As pointed out somewhere artwork (in fact most things you are going to study close up) may not stitch correctly. Much better to have the artwork horizontal or vertical anyway.
-
@ Bob -
Where the heck were you back 5 years ago when I started this? You could've saved me a ton of work. ;)
Seriously, though, where I live, this is a huge issue. I shoot for artists, museums, collectors... the whole impetus for this project was one of my best clients simply not being happy with results from traditional tools and methods. (Read: medium format backs, Betterlight, and, pretty much the top of the heap, the Cruse Scanner, used by no less than museums like the Getty. I was a dealer. I've used them all.) When I have a good client who's unhappy with the results from the Cruse, I have to assume there's a problem. Working with him, and pressing a lot of the 30+ years experience I have with shooting paintings into play, we realized this is the solution... and, as far as I've been able to see, the only solution. Shoot a small frame, light it properly.
I've written a ton about it on my blog, feel free to check it out: https://teddillard.wordpress.com/category/fine-art-reproduction/ I just bought a new painting that would be a great example of how lighting affects rendering of brush stroke... if I have time I'll post examples for you, though on my blog I've already done that and more.
A couple of notes. If stitching introduces distortion, you're doing it wrong. Resolution isn't even part of the issue. It's all about the lighting. "Even" lighting misses the entire point. When you're lighting a painting, you're lighting a three-dimensional subject.
-
When you're lighting a painting, you're lighting a three-dimensional subject.
When you are lighting most things they are three dimensional. (:-) I spend ages trying to make the print look three dimensional too.
If it provides the solution you need then it is worthwhile. The customer is always right as long as they have the money.
I will check out your blog.
-
When you are lighting most things they are three dimensional. (:-) I spend ages trying to make the print look three dimensional too.
Exactly. So, for the sake of discussion, let's say I want to light a painting with a single light, to emphasize the depth and texture of the brush stroke. I want a simple shadow, very little fill. I want this because I want the print to have depth and dimension.
As you undoubtedly know from lighting any dimensional subject, if you add light sources, you're multiplying shadows, right? 4 lights, 4 sets of shadows.
Lighting a painting with a single light source is not a problem, when you use a small frame area and step the painting. A spot, say 6' away, covers a 16x20" field with no problem, and a field-evening software like Equalight balances it out. I have brush strokes with simple, distinct shadows, realistic highlights, as the artist saw them. I take several samples of the painting, and my stitching software goes a little further to even out the field as well.
The result, regardless of the size of the painting (and I've done this with well over 60" paintings) is an entire painting with the surface lit perfectly - using just one light source. Virtually every brush stroke is lit identically throughout the work. Lighting a large painting with a single light source without stepping the painting is near impossible.
That goes a long way to helping your print look 3-dimensional.
-
Ok, you have my attention. I've read the blog.
We seem to have stepped from the technicality of doing the reproduction into the creative side of the art.
to emphasize the depth and texture of the brush stroke.
If I said to most artists I know that I wanted to emphasise their brush strokes then they would probably say that if there is any emphasising to be done, they will do it.
So not sure about some of this.
The colour of the light in the artists studio or direction it was coming from or how they felt at the time is not something (I feel) that the photographer can concern themselves with. We can only reproduce what is there on the canvas. I can only measure the colour that is there (indirectly using a Colour CheckR) and reproduce that. If I do that accurately then side by side with the original, in any light, they will look the same.
The same goes for shadows. If the original and the reproduction are side by side in the gallery they should look the same. Every gallery will have lights coming from above, and so the shadows should be below, or not at all. Also there will be more light at the top if it is a large painting. It will also slope forwards. If the light is from above and the introduced shadows are from the side then it will look wrong. This is just an observation of reality, not a solution. Obviously if the artist has painted in a shadow then an additional shadow elsewhere would also look wrong.
So, what to do about shadows? I would say a very slight shadow below, that is, light from above, or no shadow at all.
4 lights don't necessarily make 4 shadows. If the lights face each other and are of equal power then there will be no shadow. I generally use a clam shell type arrangement of long strip boxes where each light is at right angles to the painting and they face each other. You can adjust the power ratio to vary the shadow.
Anyway, good luck.
-
Due respect, but this: "If the lights face each other and are of equal power then there will be no shadow." ? No, and that's Lighting 101. I'd say if you think this is true, try lighting a pingpong ball with 4 lights and show me where the shadows go away...
You're missing my point on the brush stroke. Sure, the artist gives it the texture they want, that's exactly the point. My job is to reproduce that texture in a print - no more or less. Here are two examples - shot with standard "even" side lighting, and shot with the artist's studio lighting:
(https://teddillard.files.wordpress.com/2013/04/copylighting_detail2.jpg?w=480&h=311)
(https://teddillard.files.wordpress.com/2013/04/atelierlighting_detail2.jpg?w=480&h=300)
To see the full sized images go to the blog post, here: https://teddillard.wordpress.com/category/fine-art-reproduction/
Notice, in particular, that vertical brush line in the tan/beige tone in the center. The top image, lit traditionally and "technically correct", doesn't read the line of the stroke. The second shot defines it as the artist saw it when he painted it. It's subtle, I'll grant that, but it's important. If you look in other areas you'll see strokes and paint you can't even detect in the first example.
Anyway, I've done the science on the color on my blog (and in several years of research and testing - I'm an X-Rite "Coloratti", or at least was at one time... worked pretty closely with members of that team. http://blog.xritephoto.com/2012/06/ted-dillard-light-sources-color-management/#sthash.Vd20So0F.dpbs). Again, respectfully, you're missing the point about that, as well as in the quality of the shadow and lighting. As simply put as I know how to say it, if you light a subject with green light, and only green light, how can you expect it to reflect red? The subject will only reflect the wavelengths you give it, though you can try to fabricate a response with a profile, as you are forced to do with "full spectrum" fluorescent. I use as full a spectrum as I can - daylight with tungsten added - because without a full spectrum, I don't get a full palette of actual colors to print.
As far as shadows go, interestingly, in a North Light Studio the shadows under each stroke typically move to very cool, if not blue. You can see that pretty clearly with a white paint stroke, naturally. ...but, point being, I can make the shadows do whatever I feel they need to be doing, in color as well as values.
I'd suggest you do an few tests, since no amount of discussion really convinces anyone. Well, it never convinced me, anyway... ;)
Shoot a painting with a good, heavy texture paint application. To take the x-y easel out of the equation, let's use a small painting.
1) Light it with standard, two light copy lights, with or without polarizers, whatever your preference.
2) Shoot another version with 4 lights, again, standard copy arrangement.
3) Shoot it with a daylight, North light studio with not more than two down-spots, preferably halogen. Make sure you process using your Passport system, it handles that mix great.
4) For the sake of some real interesting discussion, shoot the piece with LED lighting, pick your poison as to the lighting technique - then look at that file's histogram compared to the histogram from #3 - look at all 3 channels. (https://www.youtube.com/watch?v=QcpGsLhp6K0)
Shoot, process, and look at the files at "Print Size". Now look at the images zoomed in until you can see the detail of the brush strokes and shadows on them.
Make some prints. Note the differences between the color range with the North Light studio lighting, particularly with a painting with difficult, gamut-stretching color. If possible, show the prints to the artist and see what they like best. Come to your own conclusions...
(Yep, it's a lot of work, I know because I've done all of that, for several years, as well as running some actual science. :) I've also proven out the "theory" with about 5 years of using the method and hearing artists' reactions... that's when I hear about the photographers they used to use, who "never really got it right". My pricing, by the way, is no more expensive than most lab's standard "high resolution" capture.)
My conclusions are that a North Light studio emulates the lighting that the painting most often is viewed under, both in quality and in color. Virtually every gallery I've ever been to has a combination of halogen spots and ambient daylight fill. Most artists that I work with work under similar conditions, with ambient daylight, and work lights or halogen spots - even when they say they have a North Light studio, they still add the task lighting. If I shoot that way, my lighting looks familiar to the artist, and the color of the capture is easier to print.
---back to the x-y easel, now ---
Now, let's assume you agree, and want to use that lighting formula on a large, 3' x 5' painting. How would you do it? If you try to light the whole work like that, you're going to get huge drop-off from your fill - maybe up to an entire stop, depending on how far you are from the North "light". No single halogen bulb I know of is going to light that area evenly, or even within a stop, from the center of the hot spot to the fringe. If you add more lights, you add more shadows, and the stroke and texture gets confused.
Thus, the 16 x 20 (actually 24", with the D800) frame. I can light it correctly, and "apply" that lighting over the entire surface of the painting, by stepping the work and not the camera.
As a maybe kind of interesting aside, watch how artists work some time. Though they do, in most cases, step back and appraise the work from a distance, they're generally working within arm's length of a painting. Their field of vision is - you got it - about a 16 x 20 frame. They move the work, on the easel or the studio floor, around so that the area they're working on falls under the task lighting they're using. For all intents and purposes, their field of view and work area is coincidentally just about identical to my capture area, both in size and lighting treatment. Again, I'm making a capture exactly matching how they've looked at their work throughout it's creation. This may seem academic, but first, it's been demonstrated again and again, and, second, guess what distance most artists end up evaluating my full-scale prints from? That's right. Arm's length, no matter the size.
** Note: I'm about ready to wrap up the prototype easel and move it into the studio for testing against my first motorized version, so I'm thinking this thread may have run it's course. I'm hesitant to start a thread about the entire technique, since I've spent countless hours documenting the process over the years on my site, along with some videos et al, but if there's interest, maybe we should do that in another thread? I'll leave it up to y'all, but I'm not sure this is the category for the discussion...
Thanks for the interest!
-
aaaaaand we're ready for the studio! ;D
(https://scontent-iad3-1.xx.fbcdn.net/v/t1.0-9/13590371_10210277967525104_8893700922798289737_n.jpg?oh=1a2ef41bcb3ae00509c08a1c90e4335b&oe=58332FBC)
-
I've had a chance to do some shooting today, and what walks in the door but a 60" tall painting. The math works - using the D800 vertically, it tracks out to 60" no problem.
...and the thing works AMAZINGLY well, if I do say so myself. ;)
-
A few shots of the final configuration in the studio, in use:
The 12V DeWalt battery power supply, motor control and drive wheel:
(https://scontent-lga3-1.xx.fbcdn.net/v/t1.0-9/13606991_10210306926009048_1105006981929322972_n.jpg?oh=692e43c10b6b03317da0f3af1568bfe9&oe=57EC0D19)
This is the bottom clamp tray for the painting - extra-wide, several mounting possibilities, along with a little storage tray (for the Colorchecker, of course :) :
(https://scontent-lga3-1.xx.fbcdn.net/v/t1.0-9/13680598_10210306925329031_4766386033339477830_n.jpg?oh=f1943a138f5fff0471fc373fdf4730db&oe=582AF6E8)
The side standard, showing the magnetic scale on the side, the vertical drive screw and the position of the power supply:
(https://scontent-lga3-1.xx.fbcdn.net/v/t1.0-9/13654275_10210306925369032_8545417180645443701_n.jpg?oh=5d3654b20ecaa514be92ae29f0c3cce1&oe=57FAFA94)
This is the screw assembly for the vertical rise - pay no attention to the blade burns on the angle braces, I didn't feel like swapping out to a rip blade that day. :o :
(https://scontent-lga3-1.xx.fbcdn.net/v/t1.0-9/13600076_10210306925409033_8241749062634803622_n.jpg?oh=ad1b539c77266f2cb5aa8c5f5230d4c0&oe=583647F9)
This shows the dual rail assembly, which is what allows an 80" canvas:
(https://scontent-lga3-1.xx.fbcdn.net/v/t1.0-9/13692507_10210306925289030_8902550141985178326_n.jpg?oh=fbd192c5c216f8cd8888cab9fa6d8e2c&oe=57F8B847)
At this point, about the only thing I can think to add would be a shutter release button on the remote. 8)
-
OK, folks - time to step off the thread. I've put it through all the testing and it rocks! ...better than I dared hope. I'll try to check in with this thread, but I can't promise I'll be looking regularly.
I've updated the site to include the "Servo" model: https://xyeasel.wordpress.com/
Please feel free to contact me through the site, or message me here if you're interested. I'm working on pricing and delivery, and will update the site with those details as soon as possible.
Thanks for the input, and keep those cards and letters coming!
update: I just started a new thread over here that shows the updated, legit version: http://forum.luminous-landscape.com/index.php?topic=114871.0