Use of Video to create stacks

Here are links to articles for your reading pleasure. You may also submit brief reviews or discuss the contents of the articles.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Clintonwake
Posts: 216
Joined: Wed Feb 13, 2013 5:49 am
Location: Australia

Use of Video to create stacks

Post by Clintonwake »

Hi All

FYI and comment, I am just recently started using Magic Lantern and doing some testing on using video to create stacks from the movie frames. I have also tested silent mode but this test is using video. Once I understand it enough and am confident to repeat the tests with "acceptable" results and also document the workflow, then I will publish my findings and share for those who are interested. I tested on both Helicon and Zerene more for curiosity but the test is to determine if I can improve my workflow/productivity and get better/improved results without using actuations .

2 tests carried out, movies ranged from 3gb to 15gb and translated to 1056 pics and 3719 pics respectively. I will keep you informed of my progress, The results are acceptable in my humble opinion however I am new to video and have not worked out how to get the optimum settings/light and the rail travel speed on the automated rail or doing it manually.

Image

Image

JH
Posts: 1307
Joined: Sat Mar 09, 2013 9:46 am
Location: Vallentuna, Stockholm, Sweden
Contact:

Post by JH »

Hi Clintonwake

Thanks for posting this – this is really interesting.

What is the pixel size of the single pictures in your stack? (I have never used Magic Lantern so excuse me if this is a stupid question)

I have stacked the pictures from a video film from a Canon 6D. The reason I did this was that I had a live subject, so I did not have another option. Result was ok, the speed of capturing the pictures was extremely short, the shutter could rest, BUT the resolution was in my taste too low to replace my normal workflow for dead subjects. What I really would like to see is a stack from a 4K video.

Due to the fact that it is possible to take so many pictures with video I also wonder if someone knows if there is a possibility to increase the resolution by blending several pictures. The normal technology for this – if I understands this right – uses a sensor that shifts in X/Y axis. I do not have the knowledge to know if there is the same possibility shifting the sensor in Z using smaller increments than we normally uses for stacking.

Regards Jörgen

Clintonwake
Posts: 216
Joined: Wed Feb 13, 2013 5:49 am
Location: Australia

Post by Clintonwake »

Thanks, these are my initial tests.

I was using canon 5dmkIII, MPE65, iso50, 1/100, f5.6 at about 3x. With Magic Lantern, I was able to set the FPS to about 24fps but with a hack as follows: I set the resolution to 1920 x1080, then set the LCD on the camera to 5x by clicking the enlarge button. This enable be to get just past the 2k resolution. The challenge is spreading that 24 frames/sec equally as the rail is moving.

I then took the video at 2k and extracted the raw frames. I then upscaled each image from 2k to 4k in a RAW editor. There was no impact to quality, in my opinion. Images were then edited very lightly in raw editor eg. exposure, noise etc and saved. each image becomes about 40mb , 4048x 2000 approx which is not bad.

The issue that I need to work out is how to get SMOOTH and CONSISTENT movement plus a smooth video ( correct settings) to ensure that no frames are missed. I will keep you posted

Post Reply Previous topicNext topic