I need to create animations for Deep Learning

Hi there,

For my project, I need to create realistic animations and render scenes into pngs. The next step, I have to mask objects in the scene so that I can do semantic labeling for training purpose.

The problem is, I cannot find an easy way to create realistic scenes and render it out at 25fps.

Do you have any idea? Which way should I go exactly, what would be the easiest way?

So again, there will be objects (i.e., actors(human-like) will go back and forth in a corridor) and a fixed camera will be watching them (i.e., video surveillance). I will need to render images at a rate around 20-25 fps so that I can use those images to train a convolutional neural network.

Thanks