here's what I wanted to do:

I'm manually rotoscoping a 4 seconds animation. I'm only tracing black lines over transparent background (png format)

I had a script in processing that read images and created cubes where black pixels were found (I have also a similar script that creates cubes depending on the brightness of the pixel read). This was done in processing using the nervous system library to export the results to obj files. That library does not work anymore with Processing 3, so I thought I could try to do a similar script directly in blender.

I want the script to repeat the steps for every frame I've rotoscoped. so here my logic 1. open image file 2. get image pixels 3. create cube where black pixels are found (or if a color image is used, create a cube depending on the brightness of the pixel) 4. save file with all cubes in place 5. close image and open next one in the sequence.

Once I get all the blender files (cubes in the shape of rotoscoped body), I want to run the models in sequence to watch the rotoscoped animation in 3D.

maybe animation nodes is the answer, maybe coding...not sure. I'm pretty new (and I'm not really a coder).

  • $\begingroup$ Adding cubes and even tracking over movie run time is trivial in Blender, but auto assigning markers (tracking elements) to the image then collapsing into a single roto-ed object is not. Usually rotoscoping requires that we break down a complex shape into gross parts, then pivot them at joints. This reduces the potential for vertex crawl where many points can bunch up along an edge during tracking. $\endgroup$ – 3pointedit May 9 '19 at 5:53

Your Answer

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

Browse other questions tagged or ask your own question.