Lionel Kobi
Forum Replies Created
-
I’d use the first part of the tutorial to match simple 3d geometry to your image. You don’t have to animate the camera or distort the image if you don’t need to.
Once you have 3d geometry matched from your camera’s perspective, it should be easier and more accurate to place objects in your scene. Having the correct focal length here should help, especially on extremely wide lenses.
The reason I think there is no one click solution for this is that a 3d match moving program requires some motion in order to solve a camera using optical flow. An image simply doesn’t provide this information.
-
Depending on how complex your effects work is, you could do the compositing using the H264 video file and once you are finished, collect the files and send the After Effects project to your editor. They would then replace the H264 file with the ProRes file and render out the comp. I have done this successfully, however I was able to sit with my editor when replacing the footage.
Also, motion tracking with H264 can be quirky. Best to use an image sequence, however you might still get compression artifacts which interfere with your tracker.
-
You can change the camera units in After Effects, just double click on your comp’s camera and change the units to mm on the bottom left.
I would use the 3d camera projection technique to match your camera, although this is still “eyeballing” to some degree. An excellent tutorial on this can be found on VCP (https://www.videocopilot.net/tutorials/3d_camera_projection/)
-
By “final animation” do you mean once you have rendered out your animation using the render queue, your image changes?
The image you posted (https://images.creativecow.net/231135/4-1-20122-23-16pm.png#14588), looks to me like a trial version of a plugin is rendering a watermark into your animation, especially the two red lines on either side.
Can you give us more detail? What plugins you are using, your other layers, etc.