I never used Topaz before. But I’m in the last stage to finalize a short animation film.
The images were made with Clip Studio Paint, all in Bitmap (not with vector graphics), all 4K
There are a few scenes, where I want to Zoom in, maybe 50 %. I guess I will then get into troubles when the film is shown on a big screen.
My question is whether I should put the video into Topaz Video to do this?
As it is a frame by frame animation, I also have the single pictures. Would it be better in terms of quality to put the single pictures into Topaz Photo instead?
They’re talking about temporal consistency. Here’s an AI explanation:
In AI video upscaling, temporal consistency refers to the ability of an AI model to maintain stability, coherence, and smoothness across consecutive frames in a video sequence.
While a still-image upscaler only needs to make one picture look good, a video upscaler must ensure that the generated details (textures, edges, lighting) remain identical from frame 1 to frame 100. Without it, the video will suffer from severe flickering, “shimmering,” or “morphing” artifacts, where details constantly change, melt, or drift, ruining the visual quality.
Whether that’s a big issue for your animation depends on its content.
It means that the neighboring frames are not flickering in the little details and the algorithm takes into consideration n frames before and after the frame (usually n=8 frames or more).
If you upscale every frame independently, you’ll see the little differences between frames.
There’s a good new model for Topaz Video - Proteus Natural.
It can upscale your animation without adding unnecessary details. You can send me a short 10-seconds clip and I’ll upscale it for you to see the difference between Topaz Photo and Topaz Video in different models.
