Hello Support Team,
I’d like to suggest a feature that could significantly enhance your software and help many video producers globally. In video production, workflows often involve dealing with compressed footage, which introduces artifacts such as pixel variations, visual distortions, and data loss (example video shown below).
Please note, I’ve attempted to address this issue using your denoising features, but they didn’t resolve the artifacts. I believe the AI is understandably trained to target actual noise rather than artifacts, which seem to be completely ignored.
I also tested various settings under Theia - High Fidelity & Details, including the “Fix Compression” feature. While it helps to some extent, it doesn’t seem designed to target red artifact blocks or other compression issues. If there were a dedicated feature to specifically address artifact removal (targeting red artifact blocks, banding, and more), it would be incredibly useful.
In my case, I made an error during video encoding, overcompressing some footage in my project (example video below). This resulted in the artifacts I mentioned earlier. Despite this, the videos remain of high quality, which leads me to believe AI could restore the lost details by analyzing other areas of the frames and recovering the data.
If any software could achieve this, it’s likely Topaz Video AI, which I already use and appreciate for its sharpening and focus-recovery tools. The only issue I’m facing in this project is the overcompression mistake, but I believe Video AI has the potential to correct it. I’m sure many other video producers would find this feature invaluable.
I’d appreciate it if you could let me know if this feature is in development or planned for the near future. Many of us would greatly benefit from it. Thank you!
• Video explanation link with examples: Compression Artifact Explanation.mov - Google Drive
Best,
Ben Rowe