View the video: https://youtu.be/UZOcnwywWhU
There’s’ a Major Update to the best Video Enhancer - Topaz Ai 5.2.1 which has 3 exciting new features
A new Rhea Enhancement model, an After Effects Plugin for frame interpolation, as well as Alpha layer support in Topaz Video AI and finally a new Pro level licensing arrangement
In this quick video I’m going’ to concentrate on the new Rhea model, which in effect is a combination of the Proteus and Iris models combined.
I’ll quickly cover the other new features at the end of the video.
The intention for the new Rhea model is to retain fine details without being destructive to text in the videos
The outstanding and unique feature of Rhea is that it processes the enhancement at 4x scale and then downscales to your selected output resolution, this comes at a cost though, the processing time is considerably increased.
In this example I am using a 960x540 image and as you can see in the input on the right hand panel
I’ve’ set the output to 7680 x 4320 - that’s 8K
I’ve set the model to the new Rhea.
You’ll notice that there is a warning about the extra resources required.
I can confirm that the processing took much longer that the single Proteus model did - but look at these results - outstanding.
At the end of the day, if you need this sort of enhancement you should be prepared to spend that extra time.
Also part of the 5.2 release of Topaz 5.2 is an After Effects plug in that works directly in Adobe After Effects and includes access to all Frame Interpolation models for up to a 16x slow motion conversion.
When you are in After Effects, you need to set a new composition for the new slow motion clip
Place a “Null Object” layer on top of the input video, the length of what the clip will be after slow motion is created.
If you want 3 x slow motion on a 2 second clip allowing you would require a Null Object layer the length of 6 seconds
Then pre-compose the two layers and apply “Frame Interpolation” from the Effects panel.
The third new feature of Topaz Video AI 5.2 is Alpha (transparency) layer support.
Video AI can now copy alpha layers from inputs and merge them back with AI model output videos.
You will need to use, QuickTime Animation, TIFF, PNG and image formats for the Alpha layer to work.
Finally for larger organizations.
As part of the 5.2 release Topaz now have a new Pro level licensing option for large professional organizations.