5

I used to use Camtasia on Windows for editing videos, specifically recorded screencasts. One thing this did well was give me the ability to draw simple annotations on top of the video, including canned symbols like arrows, along with text boxes that I could enter text into.

I was hoping to find something on Linux that did the same thing. I've tried pitivi and openshot, and looked at the docs for kdenlive. None of them can do that.

So then I figured that the next possibility would be to extract an individual image from the video, then use something like gimp or some other drawing program to draw my annotations onto a separate layer on top of the image, then extract that layer to an image that I can then import into one of the video editors and place it at the correct point in the video.

I believe I've seen some command-line tools that generate individual frame images from a video using a timestamp range. Would something like this be part of the process?

Has anyone put all of these pieces together in a real process, as opposed to just a theoretical idea?

David M. Karr
  • 891
  • 1
  • 7
  • 16
  • https://stackoverflow.com/questions/13390714/superimposing-two-videos-onto-a-static-image/13405214#13405214 might help if you are open minded on using a command line instead of a guy. – humanityANDpeace Feb 27 '16 at 08:30
  • 1
    First, I don't understand how that thread helps me. Second, "instead of a guy"? What does that mean? – David M. Karr Feb 27 '16 at 17:59
  • Please excuse the last thing is a typo, of course I meant to say GUI (graphical user interface), which contrast a command line interface. The thread features the question where somebody uses the `ffmpeg` tool to create two videos on top of an image. I did not say it is the solution, but when reading your question was led to believe that such a thing is quite close to what you asked namely drawing on top of a video. If it is not helpful to you I am sorry. Good luck :) – humanityANDpeace Feb 27 '16 at 18:43
  • The problem is, I have to find a way to combine these myriad solutions to develop and actual working process. I can use Pitivi to find the timestamp that I care about. I can use ffmpeg to generate individual frame images (although I'd like to figure out how to make it generate just a single one instead of hundreds, even for a millisecond range). I can use "shutter" to draw arrows on the image (and likely text), but that tool doesn't appear to let me separate layers, so that's useless. – David M. Karr Feb 27 '16 at 23:18

1 Answers1

4

It's possible using Kdenlive and GIMP.

In Kdenlive, put your video clip on the timeline in the 2nd video track. Scrub to the frame you want, then right click on the project monitor and select 'extract frame'. Save it, then open that file in GIMP.

In GIMP, create a new layer and draw your annotation, then hide the background image layer and, from the file menu, select 'overwrite ' to save it.

Back in Kdnelive, import the png annotation (project > add clip), and place it in the desired position on the 1st video track. If you're using the 15.12 version of Kdenlive, that's it. On earlier versions, you may have to add a composite transition for the duration of the annotation to preserve the transparency of the png.

magicalex
  • 141
  • 3