- Time
- (Edited)
- Post link
little test i did using AI based upscaling.
https://www.youtube.com/watch?v=WgnPI3zYNz4&feature=youtu.be
software: https://topazlabs.com/gigapixel-ai/
Topaz Gigapixel upscale of the GOUT - a test & info project
- Time
- Post link
I don’t remember the GOUT being that compressed…are you working with the source file or a re-encoded RIP?
- Time
- Post link
I have never actually watched the GOUT dvds, but it looks like the upscaler would work for bringing out details in elements, not whole shots.
- Time
- Post link
I don’t remember the GOUT being that compressed…are you working with the source file or a re-encoded RIP?
re-encoded rip yeah, it could also be youtubes compression or my render settings in sony vegas
- Time
- Post link
I don’t remember the GOUT being that compressed…are you working with the source file or a re-encoded RIP?
re-encoded rip yeah, it could also be youtubes compression or my render settings in sony vegas
I’ve been doing some work with Gigapixel on some of my upscale projects. Did you use regular old gigapixel and do a frame by frame or did you try their beta video upscale software?
- Time
- Post link
just regular old frame by frame gigapixel
- Time
- (Edited)
- Post link
Hi OldMateMatt,
What are the settings that you use (noise, blur, face refinement)?
I use GP a lot and I have found that the main problem in frame by frame processed videos is with the stars : they are “split” (GigaPixel redraws two sharpened stars instead of one blurred) and they move from frame to frame.
It seems to me that you also have this problem in your video. Did you achieve to reduce or solve it?
- Time
- Post link
just regular old frame by frame gigapixel
That seems to work best. I tried their new beta video software and it didn’t seem to yield nearly as good results.
- Time
- Post link
Hi OldMateMatt,
What are the settings that you use (noise, blur, face refinement)?
I use GP a lot and I have found that the main problem in frame by frame processed videos is with the stars : they are “split” (GigaPixel redraws two sharpened stars instead of one blurred) and they move from frame to frame.It seems to me that you also have this problem in your video. Did you achieve to reduce or solve it?
tbh i didnt even notice that issue and i forget the settings i used in gigapixel.
- Time
- Post link
Hi OldMateMatt,
What are the settings that you use (noise, blur, face refinement)?
I use GP a lot and I have found that the main problem in frame by frame processed videos is with the stars : they are “split” (GigaPixel redraws two sharpened stars instead of one blurred) and they move from frame to frame.It seems to me that you also have this problem in your video. Did you achieve to reduce or solve it?
I know exactly what you are talking about. I personally have found the best results are on low/low anything more than that just amplifies the mistakes Gigapixel is making. I’ve been thinking about doing a double pass, at low/low and then just a plain scale with no image improvement, and layering the two images to try and reduce the enhanced noise/small details that become distorted.
- Time
- (Edited)
- Post link
Hi OldMateMatt,
What are the settings that you use (noise, blur, face refinement)?
I use GP a lot and I have found that the main problem in frame by frame processed videos is with the stars : they are “split” (GigaPixel redraws two sharpened stars instead of one blurred) and they move from frame to frame.It seems to me that you also have this problem in your video. Did you achieve to reduce or solve it?
I know exactly what you are talking about. I personally have found the best results are on low/low anything more than that just amplifies the mistakes Gigapixel is making. I’ve been thinking about doing a double pass, at low/low and then just a plain scale with no image improvement, and layering the two images to try and reduce the enhanced noise/small details that become distorted.
Hi riftamos,
Thank you very much for sharing your tips!
Did you try doing a double pass, or did you just think about doing it?
On my side, I tried to increase the original frame with a simple bilinear resizing (ia from 720x480 to 960x640) before upscaling it with GP (ia from 960x460 to 1920x1280): that reduces the distorstion of small details, but that reduces also the advantage of using GP…
- Time
- (Edited)
- Post link
Hi OldMateMatt,
What are the settings that you use (noise, blur, face refinement)?
I use GP a lot and I have found that the main problem in frame by frame processed videos is with the stars : they are “split” (GigaPixel redraws two sharpened stars instead of one blurred) and they move from frame to frame.It seems to me that you also have this problem in your video. Did you achieve to reduce or solve it?
I know exactly what you are talking about. I personally have found the best results are on low/low anything more than that just amplifies the mistakes Gigapixel is making. I’ve been thinking about doing a double pass, at low/low and then just a plain scale with no image improvement, and layering the two images to try and reduce the enhanced noise/small details that become distorted.
Hi riftamos,
Thank you very much for sharing your tips!
Did you try doing a double pass, or did you just think about doing it?On my side, I tried to increase the original frame with a simple bilinear resizing (ia from 720x480 to 960x640) before upscaling it with GP (ia from 960x460 to 1920x1280): that reduces the distorstion of small details, but that reduces also the advantage of using GP…
I have not had a chance to try the double pass & layering of the frames yet, but I will report back when I have had a chance to try it. I was also thinking about trying a single pass on low/low then running a separate pass just as a resize with no filter and using TOOT to average between the two - generating a new frame (as compared to laying two frames on top of one another).
There is a brief discussion about TOOT over in my “Faces” upscale thread.
One thing that I have tried and I know works, but is incredibly cumbersome is “masking.” To me, it seems like GP is great on larger objects that are “up close”/the foreground in the frame but it makes a lot of mistakes on small details in the backgrounds. I’ve successfully made a mask to show just the enhanced detail of the foreground images overlayed over an enlarged background without enhancements - its not great, but its a step in the right direction. It takes a lot of time and the more elements there are in a frame(or scene) the harder it is to pull it off.
What would really be ideal is more control over where GP focuses during its upscale when its attempting to add in details. I’m also giving “Phoenix” a try which could be promising.
- Time
- Post link
For those that are curious -
I found that for my purposes, a frame by frame scale in the regular gigapixel software generated better results than using their beta video upscaling software.
I have tried a double pass and I cannot recommend it. I have also tried to layer a pass of low/low on top of a pass of none/none - my feelings on this are mixed, depending on your software you could adjust the opacity of the layers until you find your desired/optimized results.
For me, I’m finding my best results scaling with the none/none preset. It does actually do a fair job of preserving detail and removing a layer of blur even on the “none” preset. At most I run a low/low.
I think with certain scenes you could play around with a variety of filters and use masking to pic out the best parts of the best runs. The background of space for example while inside the x-wings will look better with a none/none, but Lukes face will likely look better at low/low. Use the none/none as a matte and mask Luke back into the frame by hand.
- Time
- Post link
For those that are curious -
I found that for my purposes, a frame by frame scale in the regular gigapixel software generated better results than using their beta video upscaling software.
GP has released a new app called Video Enhance AI. It is so much better than GPAI beta.
Check it out. I have posted a few comparison iamges of my Beauty and the Beast LD at the following thread:
- Time
- Post link
For those that are curious -
I found that for my purposes, a frame by frame scale in the regular gigapixel software generated better results than using their beta video upscaling software.
GP has released a new app called Video Enhance AI. It is so much better than GPAI beta.
Check it out. I have posted a few comparison iamges of my Beauty and the Beast LD at the following thread:
I was just playing with it the other day, definitely a huge improvement. From my experience, I still seem to be getting better results doing a frame by frame in just regular old gigapixel, but I still need to explore the settings in Enhance more.