Messages

Request:
Assign a custom 1D or 3D LUT to the cameras Live View display (without baking-in any actual color transformation into the video file), to allow the photographer/cinematographer to get a live (but rough) idea of what the shot will look like after color correction.

I mentioned this awhile ago and there seemed to be interested, but I haven't had the time to try it.  Is anyone else up for implementing it?

comp artists and colorists normally have a fast way to disable/enable a LUT (nuke & shake).  if there is a way to assign toggling the LUT on/off to a hardware button that would be great.

is there a measured documented device color profile for the canon's camera LCDs?  It could help provide a more accurate LUT overview, if we incorporated some kind of color management....just dreaming.

Great point.  They do exactly that!  Where LUTs would improve upon that would be that LUTs are an industry tool used in many different pieces of software, where as picture styles are only relevant to canon cameras.

What LUT previewing on your LCD screen would allow:  You could grade a shot in a popular color editing suite then export the LUT of the grade to the camera.  Once on the camera, you could preview the color of the shot (roughly) while recording, as it will appear once you apply the color correction to it.  You'll need similar lighting, but would be extremely helpful.

right.  I was a bit vague.  It would be wonderful if I could use a 1D or 3D LUT while previewing/capturing footage on my cameras LCD screen.  Nothing that would effect the video, just the preview/playback when enabled.  I could get a rough idea of how my light would be effected with the transforms from my color grade.  Perhaps its a bit much to press on the digic procs, but I had hoped it would be worth checking into.

It would be nice to be able to switch back and forth in realtime with a LUT exported from Da vinci.

IS this working on 60D yet?..... I know you guys have been busy giving the camera industry a bit of a shock..... anyway, I tried the RAW histogram with no success, perhaps I missed something obvious.

I'm not familiar with the final solution, so please excuse my ignorance if I'm asking a question with an obvious answer to most of you? Has anyone tried using Optical Flow algorithms to reduce the flicker?  Or is this using optical flow?

Your background and passions in life that have lead you here matter.  Some are members here because they enjoy coding, others enjoy photography, others enjoy tech in general.  Some are professionals in fields related to photography or where photography is a skill set used daily, others are amateurs who love photography and the digital advancements made in the dslr field or some might be trying to build a dev. portfolio and using ML as an open source project to gain knowledge and experience through. 

Either way, I'd be interested in hearing your connection to photography, background and reason for being part of the ML community.

I see people make statements that I think of as shortsighted, but if I look at it from a different point of view,I get a better understanding of why someone perhaps has a different value for a feature in ML than myself.  I thought this thread might help everyone get a better understanding for each other as well.

I'll start:  Profossional photographer on the weekends, shooting 8-9 years.
professional Background: compositing & 3D VFX
Reason for joining magic lantern community:  Wanted to assist in feature requests that are relevant for post work.  Also enjoy getting the most out of my camera and moving beyond old limitations.

what they're doing isn't that ridiculous, though extremely helpful.  A lot of the ideas and technology behind the tool have been around, but just not brought together for a single tool like this.
That said, I did notice a few items that did look like artifacts, though its hard to tell if they are from the online compression or from the tool itself.  I'll have to have a look at the documentation to get a stronger grasp on it as a whole.

Adobe released a tech demo video of a similar idea they were working on several years ago.  As far as I know, it never made it into a consumer level Adobe product.  What happened to it? I'm not sure, but I imagine it wouldn't be to hard to find out the reason for the delay.

http://www.youtube.com/watch?v=1PktKqyRXIE

Ron Brinkmann (visual effects artist and digital imaging pioneer) mentions the value of a RAW histogram in his blog here http://digitalcomposting.wordpress.com/2011/05/20/your-camera-it-lies/

Alex you might be interested in his other entry here about his desired camera functions:  Later in the article its becomes a bit more software focused.
http://digitalcomposting.wordpress.com/2012/07/02/camera-features-i-want/
If you haven't gotten any attention from him or any of his associates yet (I'm sure you wouldn't be able to talk about any specifics or out of manors just not), he might be interested in what ML is doing these days.  Though I'm sure he and the rest of the educators & leaders in VFX are paying plenty of attention (if not for professional purposes now, at least for personal).

What variables determine the limit of a cameras burst rate (aside from marketing)?

Is it the build quality of the mirror and other moving parts or is it a memory throughput issue?
If it's a memory throughput issue, would it be possible to lower the resolution to allow a higher burst rate?

Some P&S cameras have "super" high speed burst modes at extremely low resolutions (for photos), quite gimmicky.  I was thinking more like getting just another 2 to 3 photos-per-second by switching to sRAW or mRAW when I shoot.

This would be great!
Most photographers can switch without thinking, but thats because they're used to the hardware.  If a decent on-the-fly software solution was taken (one button hold), modern GUI for metering modes, it could make getting some shots much easier.

several months ago I recommended something similar or if it could log the angle changes, but it didn't seem to pick up much interest.
    I agree it would be nice to start recording that data.  Paired with the zoom and focal point data already being tracked, it could be quite handy to import into a Comp project. However, I guess with how easy tracking and match moving is these days with 3D trackers, planar trackers and 2.5D tracking it didn't pick up much interest.  The issue would also be that AE, Nuke, AVID, Blender or whatever your comp app choice might be, it would have to be able to read the log file or we would have to be able to convert the log file into a generic camera transform file.

      With all the sensors slowly being added to cameras, I was hoping Magic Lantern would (over time) log enough info from cameras to moved toward a home-brewed steady-cam solution.  Not for layered motion passes, but for importing camera transformation and focal point data into comp and 3d applications.   I think it'll be awhile yet.

Not sure about usefulness in stills mode.

I agree, the practicality dramatically increases with video, yet I believe it is relevant for still photography as well.  Especially if you are documenting a build-up of some sort, over the period of a few seconds.  You don't have to wait for the first several useless photos to write to disc before you can start shooting at the burst speed again, you can just keep your finger on the shutter to let those photos get overwritten in the buffer without missing a beat.  Then let up on the shutter to allow your group of shots to get written to disc.  That could be a huge feature at the right time.

point taken.  The constant shutter sound could take a bit to get used to.  The other cameras I've seen with this feature (including high end cameras) certainly weren't DSLRs like you said, but UHD video cameras (capturing jumping great white sharks).

Perhaps I didn't explain it very well.
The camera doesn't begin capturing photos until after you've pressed the shutter button (or are holding it down), then quits taking the photos once you press the shutter again (or release the shutter) at which point it writes the last 5-24 photos (whatever the buffer can hold) to card.

This could be quit helpful for when you're working with animals or children.  If you have to get a dog or baby to do something, it can be difficult to determine when that'll happen, perhaps bad starts or "misfires" on a trick or a smile, but within a second or two they'll hit that desired look or pose.  So a continuous pre shot would be handy for that, so you wouldn't have to wait for the buffer to clear and write the photos to disc, you just keep photographing until he hits that pose, then hit the shutter button again to finally write whats left in the buffer.

I understand the idea of shutter abuse, though I did purchase my camera with the intent to use it to get the shot, even if it means pushing it to the limits.

My old nikon point and shoot from 2004 had this feature (functional even when capturing RAW at 8MP).  I'm not sure its something ML can allow, but I wonder if the hardware demands aren't that high.

I might be off base, I am quite often, but I believe the two main functions necessary for this are
1. Allowing the buffer to write over itself once its full.
2. Allow the buffer to cycle (start at the beginning, once its full)

Lowering the capture resolution would greatly increase the use of this tool, allowing more images in the buffer.  So MRaw or LRaw would probably make this more useful.

Clint

"Pre-shot" burst mode: as described by casio " a ultra-high speed continuous shooting mode that captures images from the scene prior to the moment when you actually press the shutter button"
as soon as the photographer presses the shutter button only the last user defined amount of photos are kept.

It's great if you're waiting for a once in a life-time moment to happen and don't want to miss it.  You can just fire away before the action happens, then disengage the continuous burst mode (normally a press of the shutter button) once the action you want to capture finishes, without using up your entire memory card.

If you linked this with any of the motion detection shutter controls, you'd have something very interesting.


Some high-end video cameras have this feature as well as some point and shoots.