I have used Icarus motion tracking software to combine camera motion from a video clip into Blender. So far, so good. Now I want to model, animate and render some 3d objects to composite over the video. Two questions:
What is the best method for compositing them? (The only method I’ve tried so far is keying the images together in Final Cut)
How do I render transparent shadows so that my 3d objects look like they’re casting shadows on the real floor in the video?
You could set your video as the camera background or combine the in post (blenders compositor/nodes, sequencer or Final Cut). What one is best depends if you wanna tweak more or get stuff done fast.
You need to add a plane to catch the shadow & then in it’s material settings press the OnlyShad(ows) button under Shaders. If you decide to mix in post make sure to use a format that supports alpha (like PNG & EXR).
However, I put the video as the background image in backbuf, but it doesn’t advance during render. It just stays on the first frame. How do you get the video to advance with the animation?
add an image node in the node editor, set the number of frames of your video, and add an alpha over node to put the render layer over the image(movie) node, enable composite in the anim(render) panel . Click anim !