iain.bason
Well-known member
PaPa said:
In squeeze mode it is effectively uprezing (though not to HD, of course).
How to get high quality widescreen footage
- Thread starter disjecta
- Start date
In squeeze mode it is effectively uprezing (though not to HD, of course).
Andrew Brinkhaus
Veteran
So, have we come to the conclusion that this uprez process will hardly enhance the true 16:9 image of the XL2? Is there a workaround? What about importing the 16:9 XL2 footage as 4:3, letterboxed, and then exporting as letterboxed 4:3 into Photozoom for uprez???
PaPa said:
All i'm saying is that i bet the dvx does better quality in squeeze than some weird sharpening filter + resize. And I bet it doesent do that much better quality than eg. photoshop, or virtualdub with lanzos resize.
But I'll probably have to test it myself someday :beer:
Andrew Brinkhaus
Veteran
Ok...here it is, I think I found something that works:
1) Capture XL2 footage as 16:9 widescreen footage
2)open a new 4:3 project and import those newly captured 16:9 files. "shrink" the video to fit a 4:3 monitor (should give you the whole widescreen space, with letterboxing).
3)Export as 4:3 letterbox, and begin the photozoom process as you would regularly.
Here a are a few sample grabs of resized XL2 footage w/ letterbox, ready for Photozoom! (sorry they are soft...had limited footage to take stills of)
Correct me if I am wrong, but this should yield even higher quality results than that of the DVX, simply because you are starting out with more pixels crammed into that 4:3 file with the XL2 than you are with the DVX!
1) Capture XL2 footage as 16:9 widescreen footage
2)open a new 4:3 project and import those newly captured 16:9 files. "shrink" the video to fit a 4:3 monitor (should give you the whole widescreen space, with letterboxing).
3)Export as 4:3 letterbox, and begin the photozoom process as you would regularly.
Here a are a few sample grabs of resized XL2 footage w/ letterbox, ready for Photozoom! (sorry they are soft...had limited footage to take stills of)
Correct me if I am wrong, but this should yield even higher quality results than that of the DVX, simply because you are starting out with more pixels crammed into that 4:3 file with the XL2 than you are with the DVX!
Last edited:
imageonepictures
Well-known member
Hey,
If you are shooting ona Panasonic DVX100 you could use the 16:9 adapter. This will give you true 16:9 though thus far I see no problem shooting letterbox mode.
Chris
If you are shooting ona Panasonic DVX100 you could use the 16:9 adapter. This will give you true 16:9 though thus far I see no problem shooting letterbox mode.
Chris
Andrew Brinkhaus
Veteran
Yeah, I just dont know...technically, it should enhance the footage...Its not like Photozoom's algorithm says "Your final product can only look this good." I believe that Photozoom's uprez back to 16:9 would yeild a much higher quality result...Technically, I am gaining more resolution just by shrinking the 16:9 file down to 4:3 letterbox, add the photozoom process on top of that, shouldnt it make a pretty significant difference?
rogar-Dum(DBG):
Andrew Brinkhaus
Veteran
NOT SURE. I'm trying to talk about using the photozoom uprez process with an XL2...
InvisionProductions said:
In step 1, are you capturing your footage as anamorphic widescreen? If so, then why are you letterboxing it before you uprez? That's equivalent to downsizing and then upsizing again. That seems like an uneccessary step to me, unless there's a reason for this? If you need your final footage to be letterboxed, you should letterbox after you uprez with Photozoom.
Andrew Brinkhaus
Veteran
I'm capturing just the standard DV widescreen preset in PPRO 1.5...The only reason I am letterboxing it manually is because thats how the uprez process was laid out before, and I don't want to stray from it, since it uses specific ratios and sizes...
Since I am starting with 16:9 native footage, downsizing it to 4:3 letterbox should increase the quality, slightly, correct? Since the original image has more pixels, making the image smaller should help the quality, not hurt it. Then, with that 4:3 image, export that into TIFF frames and uprez in photozoom, then reimport them back into a 16:9 project as true widescreen.
Isnt it logical that to take stock 16:9 footage, resize to 4:3, uprez using Photozooms s-spline back to 16:9, the new image would look better and higher res than that original 16:9 footage? If not, than this process is pretty much pointless. I just need to know, YES OR NO!
Since I am starting with 16:9 native footage, downsizing it to 4:3 letterbox should increase the quality, slightly, correct? Since the original image has more pixels, making the image smaller should help the quality, not hurt it. Then, with that 4:3 image, export that into TIFF frames and uprez in photozoom, then reimport them back into a 16:9 project as true widescreen.
Isnt it logical that to take stock 16:9 footage, resize to 4:3, uprez using Photozooms s-spline back to 16:9, the new image would look better and higher res than that original 16:9 footage? If not, than this process is pretty much pointless. I just need to know, YES OR NO!
iain.bason
Well-known member
InvisionProductions said:
NO.
The original 16:9 footage is as good as you're going to get, if what you want to end up with is 16:9.
Andrew Brinkhaus
Veteran
Ok, thanks. I had just figured that the Photozoom's uprezzed 16:9 would look better than the default captured 16:9. I know my 16:9 could never look as sharp as Disjecta's...
The quick answer would be "No."InvisionProductions said:
That's an oxymoron right there. Downsizing will only improve perceived quality. If you take an 640x480 picture and downsize it to 320x240, you're obviously removing pixels (i.e decreasing resolution, thus less quality). But the 320x240 picture may *look* better because it's smaller, which allows it to hide some of the flaws of the bigger picture. This is perceived quality.InvisionProductions said:
I think you're confused about what looks good to you and what has actual quality.
Photozoom doesn't work miracles. It can only do so much to improve the perceived quality of a picture. It can't add details that the original lacked.
-KiN
Andrew Brinkhaus
Veteran
Yeah, good point. I didn't think about that! Thanks for the info! Is there a way to uprez already native 16:9using the same method?
I'm in a jam...I uprezzed a 24PA shot which I set to edit ona 24p( 23.98) timeline..I uprezzed and kept in the letterbox format..when i create a Qt movie from the image sequence I created a 23.98 QT movie and a 29.97QT movie....neither will sync back to my audio..I made Qt movie of all of these, including original file and they are all different lenghts/speeds.........what do I need to do with the audio to make it sync?.............help.
Derek Botelho
Member
OK, this thread is making me cry....because I'm not so savvy. I can use FCP and do all that stuff, but I'd like to make my letterboxed 4:3 stuff anamorphic, did someone post a simple step by step guide to doing this with photozoom?
Thanks-
Derek
Thanks-
Derek
andybiz_2005
Well-known member
Hi all,
After reading all the replies on this thread, I decided to experiment on my own.
I have some 50i interlaced video footage I shot a while ago & chose to use this as my test footage.
I shot the footage in interlaced, 4:# aspect ratio & put in the letterboxing later in post (Premiere Pro 1.5. I exported the videoclip into a series of 910 .tiff frames. From there, I opened Photozoom Pro & set the resolution to 720 X 768. After Photozoom processed all 910 frames, I reimported all the uprezed frames into Premiere Pro, a new 16:9 widescreen timeline & rendered it into an MPEG2 movie. For the original 4:3 video footage, I just rendered that into a nother MPEG2 movie for comparison purposes.
The links to the two movies are at :
http://video.google.com/videoplay?docid=6738084318429997985
http://video.google.com/videoplay?docid=385784982278717932&hl=en
Strangely enough, the widescreen footage uprezzed with Photozoom Pro did not look as sharp as the original interlaced footage, perhaps because I de-interlaced the original footage before uprezzing it in Photozoom Pro. Maybe deinterlacing caused the uprezzed footage to look softer compared to the original interlaced footage.
Here are some render details. In Photozoom Pro, it took 2 seconds to render one frame.
Rendered at 11:11am
3min:41secs to render 100 frames
8min:00secs to render 200 frames
12min:05secs to render 300 frames
16min:18secs to render 400 frames
20min:34secs to render 500 frames
24min:34secs to render 600 frames
28min:40secs to render 700 frames
32min:45secs to render 800 frames
36min:54secs to render 900 frames
37min:21secs to render 910 frames
Stop rendering at 11:48am.
Total duration of movie is 35 seconds.
Used an AMD Athlon 3000+ (2.0GHz) processor with 2GB of DDRII RAM.
After reading all the replies on this thread, I decided to experiment on my own.
I have some 50i interlaced video footage I shot a while ago & chose to use this as my test footage.
I shot the footage in interlaced, 4:# aspect ratio & put in the letterboxing later in post (Premiere Pro 1.5. I exported the videoclip into a series of 910 .tiff frames. From there, I opened Photozoom Pro & set the resolution to 720 X 768. After Photozoom processed all 910 frames, I reimported all the uprezed frames into Premiere Pro, a new 16:9 widescreen timeline & rendered it into an MPEG2 movie. For the original 4:3 video footage, I just rendered that into a nother MPEG2 movie for comparison purposes.
The links to the two movies are at :
http://video.google.com/videoplay?docid=6738084318429997985
http://video.google.com/videoplay?docid=385784982278717932&hl=en
Strangely enough, the widescreen footage uprezzed with Photozoom Pro did not look as sharp as the original interlaced footage, perhaps because I de-interlaced the original footage before uprezzing it in Photozoom Pro. Maybe deinterlacing caused the uprezzed footage to look softer compared to the original interlaced footage.
Here are some render details. In Photozoom Pro, it took 2 seconds to render one frame.
Rendered at 11:11am
3min:41secs to render 100 frames
8min:00secs to render 200 frames
12min:05secs to render 300 frames
16min:18secs to render 400 frames
20min:34secs to render 500 frames
24min:34secs to render 600 frames
28min:40secs to render 700 frames
32min:45secs to render 800 frames
36min:54secs to render 900 frames
37min:21secs to render 910 frames
Stop rendering at 11:48am.
Total duration of movie is 35 seconds.
Used an AMD Athlon 3000+ (2.0GHz) processor with 2GB of DDRII RAM.
Last edited:
HaroonGhauree
Active member
Great "Poor Man's Wide-Screen " thread ...
I have a few concerns here:
Q1. Why always PhotoZoom when there is a proper HD video uprez software i.e. 'Instant HD' by RedGiant ?
Q2. If a good looking footage letter-boxed shot with a bare DVX100 (i.e. no 35mm adapter & no Anamorphic filter) not appearing so good with PhotoZoom stretch, why talking about it ? (Considering the point of debate is getting great looking Wide-Screen footage)
Q3. Why can't try other HD uprez softwares and compare their results, instead of only PhotoZoom which is taking valuable time to stretch each frame in batch and producing no better footage than the original one ?
Q4. Isn't it a possibility that 'disjecta' used some kind of 35mm adaptor to shoot or some kind of other software in the work-flow before using PhotoZoom uprez ? Coz if a good-looking footage doesn't appear as good or better in our work-flow after uprezzing there must be something missing ...
Any suggestions ...
I have a few concerns here:
Q1. Why always PhotoZoom when there is a proper HD video uprez software i.e. 'Instant HD' by RedGiant ?
Q2. If a good looking footage letter-boxed shot with a bare DVX100 (i.e. no 35mm adapter & no Anamorphic filter) not appearing so good with PhotoZoom stretch, why talking about it ? (Considering the point of debate is getting great looking Wide-Screen footage)
Q3. Why can't try other HD uprez softwares and compare their results, instead of only PhotoZoom which is taking valuable time to stretch each frame in batch and producing no better footage than the original one ?
Q4. Isn't it a possibility that 'disjecta' used some kind of 35mm adaptor to shoot or some kind of other software in the work-flow before using PhotoZoom uprez ? Coz if a good-looking footage doesn't appear as good or better in our work-flow after uprezzing there must be something missing ...
Any suggestions ...
Last edited:
