Yeah. Here’s how I would do this:
- Measure your table.
- Build a scale model of the tabletop in Blender.
- Place your camera in Blender where you think would be a good approximation of the real camera.
- Split off a new 3d window and make it a camera view.
- Go into the “buttons” mode for this window, and set it to have a background image, size 1. When you go back to the 3d view, your image should fill the extents of the camera. You may want to set the opacity of the image lighter or darker depending on your monitor, in the button view.
- Grab the tabletop in your other 3d view.
- Move it around a little, watching how the wireframe lines up with the picture in the camera viewport.
- Once you have it positioned properly, make your cube.
Then, you can either use a backbuffer in the render buttons to place your background image in the final render, OR (my preference) you can render without backbuf, but with the Key button down to an rgba targa file, then composite more carefully in your image editor of choice.