Page 1 of 1

E3D Filming Help

PostPosted: 11/30/2018, 8:39 pm
by andrejwmes
Working on filming a hovering iphone zoom effect. I have attached a mockup version below (not the final product but a base of what I am going for). I am planning on re-shooting for this effect soon and I am looking for input as to how I can film in a way to make this as clean as possible. I really just don't know how to shoot with this effect in mind. Any thoughts are greatly appreciated. Also any tips on the actual editing a more than welcome.
[youtube]https://www.youtube.com/watch?v=qThgfa-0scg&feature=youtu.be[/youtube]

Re: E3D Filming Help

PostPosted: 12/10/2018, 3:57 pm
by ctarv419
The transition from the real phone to the digital phone is pretty good, but after a couple frames of the digital phone it's really obvious that the motion is completely different. So while the actual replacement looks ok, the motion kills the illusion almost immediately. I think you need to either throw your phone intentionally in a way that's easy to match in post, or spend more time working on playing with the simulated physics of the tumbling phone to make it look more natural. Also, it seems like the zoom mechanics don't quite look right. There's no in-camera zoom, the zoom is just a scale-up of the background image right? If you really want to make it look good, you could make a note of your camera's optical characteristics, do the tilt up and the zoom in-camera, then match that move with the virtual camera in after effects. It would help to have the focal length before and after the zoom if possible. Good luck with your effects shot!

Re: E3D Filming Help

PostPosted: 12/15/2018, 11:31 am
by star+circle
I detect a conceptual problem here, which is that in reality the phone would never slow down or stop rotating unless time stopped. So, from that perspective, the phone scene should be made to look like a time freeze. The plus is that this way you don't need to replace the phone with a 3D model, you can just shoot a real phone for the zoom part and match it up at the freeze point. The freeze and the zoom would have to be a few moments apart (.05 to 1 seconds, I would say) so that the viewer has some time to realize the motion has stopped before you zoom in to the phone. ctarv419's comments about the 3D zoom as opposed to a scale apply here. Unfortunately the conceptual problem reappears at this point, since the motion on the phone screen has not stopped, so it wouldn't exactly make sense that time stopped outside but not inside the phone. But it might work, nonetheless. The viewer doesn't have as much time to figure out what's logical and what isn't. And you can always say that's the whole "trick" in the video.