I would like to see Video AI have this capability added. It seem like this would work well for old digitized VHS and camcorder videos.
I wouldn’t want it all considering it cant handle videos longer than 200 frames.
I wonder how far away we are from topaz building and being able to implement something like this.
,
Videogigagan is backed by Adobe a $214 billion company and would never be released as open source.
Topaz Labs is around the $5 million mark.
So if Adobe are no where near releasing it with all their clout I doubt Topaz could compete.
After Project Starlight maybe Topaz could look into doing VideoGigaGAN.
I realize that current hardware cannot do this, but in a couple years it may be able to.
I think implementing something complete new into a bugy program is not a good idea. What TVAI does is half fixing things and releasing new bugs, for example the VFR thing is not really solved, this and others. They have to come to a good base again first.
Starlight is really disappointing The Quality is bad, faces are distorted, it’s not much better than the older Algorithms like “Iris”…
A lot was hoped for when it was announced. But after trying it out with own clips, the Results was really more than questionable!
Its probably because starlight doing good videos which was filmed on film reel(maybe)
Waiting Starlight model for dvd/vob footages
Made another topic but I saw the comparison (of a cliff richard vhs) and here’s what I don’t understand. We all know that’s Cliff Richard. We all know what Cliff Richard looks like and so we can have a good guess at what the frames should look like,. This is information we have and information that the algo is only guessing at.
Why can’t we give the computer more information to work with.
- here is what the subject actually looks like from different angles. In high resolution pictures.
- perhaps here is a picture taken at the actual event that should show you (the algorithm) how some of the vhs distortion is working.
I have lots of home videos but I don’t want to distort pictures of my family. Why can’t I upload pictures of my family a la deepfakes to get more accurate guesswork?
Is this what gigagan means?
FaceFusion will allow you to do that.
VideoGigaGan enhances footage regardless of whether there is a face in it or not.
On the demo site there is an example of ants crawling around on the dirt.
Thanks. Does Facefusion work with video too?
Yes it will face swap photos, video and will also colorize photos and videos.