We all know how much better footage shot on a very expensive camera looks compared to a cheap camera.
How about a product which converts footage shot on cheap camera to look like $100,000 dollar camera? Wouldn’t you be able to train a product similar to your up-scaling with the same footage shot on a cheap camera and expensive camera?
I know the lighting perceived by the expensive camera would be different, but would the AI also be able to learn the correct lighting from the bad lighting?
Anthony Morganti has just done a YouTube video on this very subject!
Have you looked at Video Enhance AI?
@rayinsanmiguel Thats cool, sorry i meant specifically for video (updated thread title).
I think there technically may be a few differences between what’s required to try and match video frames to say an RED video camera (resolution aside)? I heard cameras shots inherently look better due to differences with video in exposure time, compression etc.
Just seen @AiDon’s message. Sorry I had not realized video enhance could do this. I will give the latest version a try. I was using a very old version of Topaz gigapixel so thought the benefits available were only to resolution. Not sure if perhaps it may be useful running every frame of video through Sharpen AI or is that definitely just for photography?
Thanks guys
Thanks alot
Video Enhance AI should cover all your video needs however some users do still utilize our photo apps for video in certain circumstances. Because Sharpen AI is an image product, it treats each frame like an individual still image so it could affect continuity/quality when reassembled into video and played even though the individual frames look great.
