FAQ Why you have to post-process files with ML?

Started by scrax, June 23, 2012, 02:23:58 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

scrax

Magic Lantern can do exposure or focus bracketing, HDRvideo, Silent Pic capture and timelapses.
Those functions need to be post-processed to have a workable pic or video.
For each function there is a post-process workflow so let's see them one by one:

Exposure bracketing (for HDR pics)
High Dynamic Range pics are images with an higher dynamics than what is possible to capture with our camera.
To expand the range of the camera we can take different exposure of the same scene and merge them in post.
In ML you can set how many pic to take with different exposure for each time you take a pic.
ML will then save a script that automatically converts those pic in a HDR_****.JPG when launched (needs python, numpy and PIL installed)

Focus Bracketing (Focus Stack)
When you need to extend your DOF (for example in macro shooting) you can take a series of pic with different focus point that will then be merged in post to a single higher DOF image.
In ML you can set the start and end point of the part in focus that you want for your final image ant the step sizeof focus adjustments.
ML will take the pics needed to have all in focus and saves a script that will automatically convert those pic in a FST_****.JPG

Silent Pic
ML can save the Live View image on screen to a .422 file, this is useful with timelapse to avoid using shutter.
Since .422 file are not supported a lot there is a script in the ML source code to convert those .422 file to .JPG

HDR video
Like for pic also for video you can expand the dynamic range by combining two or more differently exposed video of the same scene.
HDRvideo feature of ML consist in changing for each frame the ISO settings during a video record. This raw video will flicker if played how it is, so we need to split it in 2 video one with high exposure and one with low exposure and then we can just merge them (with some artifacts), or we can interpolate frames before merging them (this is still experimental)

Timelapse
Timelapses are video made of a sequence of photos shooted with a time interval (usually fixed). To avoid using shutter is possible to shot silent pics instead.
In ML you can set the time interval for each shoot and then you can merge those pic to a video in post.
With the FPS override option is possible to shoot timelapse video using slow FPS value (this will also save shutter)

I'm using ML2.3 for photography with:
EOS 600DML | EOS 400Dplus | EOS 5D MLbeta5- EF 100mm f/2.8 USM Macro  - EF-S 17-85mm f4-5.6 IS USM - EF 70-200mm f/4 L USM - 580EXII - OsX, PS, LR, RawTherapee, LightZone -no video experience-

nbenadom

<Exposure bracketing (fo HDR pics)
High Dynamic Range pics are images with an higher dynamics than what is possible to capture with our camera.
To expand the range of tha camera we can take different exposure of the same scene and merge them in post.
In ML you can set how many pic to take with different exposure for each time you take a pic.
ML will then save a script that automatically converts those pic in a HDR_****.JPG when launched (needs python, numpy and PIL installed)>



Ok, I'm totally new to ML. Once the bracketed images are taken through ML, how do they automatically combine? Do I just pop the SD card into the computer and they'll automatically come up already combined? I have those 3 programs installed on my MAC, but I've just been using PS and NIK HDR Efex; and with each of those, I've had to manually select the series of photos to combine.
Regards,

Noel

scrax

Hi,
to launch the script you must use Terminal
open it from /Applications/Utility you will se a window with some text
now from finder go to you card into the DCIM/10*CANON folder with the script you want to start (keep both windows visible)
then write in the Terminal window:
cd
with a space, and drag the folder
you will see the address of the folder like:
cd /Volumes/EOS_DIGITAL/DCIM/100CANON/
press enter and write:
bash
with a space, drag the script and press enter
the script will start converting the pics but remember that it support only jpg for now.
MLTools will do all this automatically when ready.
I'm using ML2.3 for photography with:
EOS 600DML | EOS 400Dplus | EOS 5D MLbeta5- EF 100mm f/2.8 USM Macro  - EF-S 17-85mm f4-5.6 IS USM - EF 70-200mm f/4 L USM - 580EXII - OsX, PS, LR, RawTherapee, LightZone -no video experience-

3pointedit

I would add ML digital ISO (DIGIC), a great option for crushing/reducing ISO (when no ND filter is available) but can result in clipping that needs to be repaired in post too.
550D on ML-roids

Urofoto


Timelapse
Timelapses are video made of a sequence of photos shooted with a time interval (usually fixed). To avoid using shutter is possible to shot silent pics instead.
In ML you can set the time interval for each shoot and then you can merge those pic to a video in post.
With the FPS override option is possible to shoot timelapse video using slow FPS value (this will also save shutter)
[/quote]


hi, I `m very exciting using this Magic Lantern firmware, but I wanna see the results of my shoots
so....i already made two examples of BULB RAMPING (sunrise and another on cloudy day), whats the next step???
As I can see most of the post says that Adobe Premiere and After effects would be work on it, to be honest i CAN'T find the way to make a timelapse properly.

cheers
i`m using Canon 5D mark II 16-35mm L f2.8, 70-200mm L f/2.8

Rehaan

Urofoto,

You need to import that sequence of still images that you took into premiere so that it makes a video sequence.

Instructions may differ for your version, but this should give you the idea:


- Ignore the part about pixel aspect ratio, yours will be sqaure.
- Do however change the frame rate depending on how fast you want it to play.

cya
R