I'm a newbie to Blender's compositor, so I don't know if this is possible or how to do it.
I want to be able to automate the process of combining multiple animation frame renders into a single image for a character turntable collage, whereby the front, front angle, side, back angle, and back renders are side-by-side as a single image. Typically, I use an external image editing program to load all the the images into a single canvas, crop them, position them, and then export the combined image. Being able to automate this process in Blender would be more efficient and accurate.
The reason I do this is to quickly compare tweaks I've made to a character between major changes as I work on it by switching between composited images using an external image viewer.
I have a single camera pointed at the character and it is parented to an empty. This empty rotates around the character for five frames for each view angle stated above. I render out the animation, and so on.