0

In my job, I need to create a video that incorporates an avatar.

The avatar will be present at the beginning of the video and possibly at the end. It will be there to give a little more life to the video but will have no other functions. A text will be said and the avatar will have to pretend to speak for me.

There are programs like Powtoon or VideoExplainer that are good, but not really in adequacy with what I seek.

So I wanted to know if there was any software where I could import an already edited video, and then create an avatar that I could add to that video. This would allow me in the long term, to have a database of avatars that I could reuse by simply importing my videos edited on other software.

If someone knows a program that does this, or comes close to it, I'm more than interested in buying it!

  • For which operating system? – unor May 31 '18 at 8:07
  • Should the avatar's lips follow the text? Or can it be any video of someone or something that talks? Is a green background OK? Do you want the text to be shown on-screen too? – Nicolas Raoul May 31 '18 at 8:30
  • For MAC. It would be better if the lips moved according to the text. Green background would be perfect. No, the text won't be shown on screen. – Guillaume Sallé May 31 '18 at 8:48
2

I have to be honest: this sounds a lot like you needing to learn a 3D (Three Dimensional) DCC (Digital Content Creation) tool, basic animation chops, and synching with the audio you have for the spoken words.

Blender, modo, maya, 3DS, C4D would all work for this. Modo is my preference, but Blender is free.

There's a recently released app (application) on the iPhone X (mobile phone & computing device manufactured by Apple) which will do decent (but not phenomenal) facial mo-cap (motion capture) using the built-in 3D camera, and you can export out an animated FBX (commonly-used 3D mesh file format - see https://en.wikipedia.org/wiki/FBX) complete with blendshapes, morphs etc.

You could create your own avatar head mesh and retarget the blend shapes and morphs from the mo-cap with some work.

| improve this answer | |
  • Please edit your answer to spell out "DCC". – Kodiologist Jun 1 '18 at 16:54
  • Thanks @Kodiologist - hopefully these edits help with clarity. – GerardFalla Jun 1 '18 at 17:19
0

I would suggest taking a look at MoviePy which runs under python 3 on many platforms behind the scenes it makes use of FFMPEG (it installs this for you on first use) & optionally ImageMagick (for text overlays you need to install this).

Assuming that you have a video clip or set of still images corresponding to the Avatar(s) making the motions for the various sound elements, (ideally one for each speech element), you can either simply insert the Avatar clip, (which can be generated to match the text), with its own background in a picture in picture manner or create a mask and overlay the Avatar on the video.

The best bit is that you can automate the process so that you can quickly create a new version for any combination of the original video and your Avatar selection.

All of the software is free, open source & cross platform.

| improve this answer | |
  • Thanks for the answer, however, the problem is I also need to create the avatar... And I would appreciate to already have templates of different avatars to choose the one that corresponds the best. – Guillaume Sallé May 31 '18 at 8:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.