Deinterlace question

should I deinterlace before or after I add any filters to a project, and does anyone have a opinion about the jes deinterlacer?

Are you trying to take 29.97 video and make it look like 24p?
If so, don't simply deinterlace - use something like Graeme Nattress's Film Look plug-in. It is considerably more sophisticated that simply dumping 1/2 of each frame.
cheers,
x

Similar Messages

  • Deinterlacing Question

    I captured some footage  as SD 60i and I wanted to create a web movie with it. I'm using Prem Pro CS3.
    Since it was for Vimeo and they seemed to want H.264, I used Adobe Media Encoder for output.
    I selected H.264, square pixels, Progressive which I thought would avoid interlacing artifacts (combing).
    However, when I viewed this in both the source and output screens one particular area of high motion (flames) both showed combing.
    I thought that progressive was supposed to remove this artifact.
    However, if I went to the original clip and right clicked - Field OPtions and set to "always deinterlace", then when I looked at the source and output screens in AME, the combing was no longer present in either source or output screens
    .  Indeed, if I rendered the clip as H.264 Progressive with field options set to None, the combing was present and if I rendered H.264 Progressive with the field options set to Always Deinterlace, there was no combing.
    Does this mean that at least in CS3, H.264 Progressive isn't really Progressive?
    Or did I miss something in the help files about having to set the field dominance to always deinterlace AND  render out as progressive?
    Thanks for your help in advance.
    John Rich

    Ann, I am trying to figure out some type of work flow for using captured AVCHD 1920x1080 60i for both regular DVD's (SD), and the web.
    Up until now, I was capturing with Cineform's HDLink and simultaneously converting the footage to Progressive.  Then editing and downsizing with Cineform and outputting with AME (CS3) to Mpeg2 for DVD and Mpeg4 for Vimeo.  In spite of being against the "conventional wisdom" of "always follow through  with your original footage settings", this seems to work well on my HDTV and PS3 I use for playing the DVD's and Vimeo.  Theoretically, no data is lost by deinterlacing.
    However, I was always careful not to pan rapidly or include any rapid action.
    From what Jeff said above, I assume that you don't lose any data if you just "create" a progressive video from an interlaced video unless you also deinterlace the video you are using to create progressive footage (ie I assume that deinterlacing discards one of the two fields per frame).  Again, from what Jeff said, if one has rapid action, then changing interlaced to progressive freezes the artifact causing slight movement between intra-field frames into each frame and there is no way to get rid of it then.
    Again, based on above, if I plan to video something where there is action, then I shouldn't convert to progressive but follow through with interlaced.  However, at the end with encoding to Mpeg4, it may be best to just deinterlace only the parts with the action to avoid artifact, before converting the footage to progressive so I would save the data in the relatively still areas of the footage (sort of crude Adaptive Deinterlacing).
    I realize this is all theoretical and it would be extra work to search out the action parts of the footage, but maybe I'll be able to work it out better in the future. Looks like I'm back to "conventional wisdom" .   Naturally, any comments from you and Jeff are always highly regarded.
    John
    ADD.  By the way, how do you output to the web with CS4?  Windows Media Encoder always has given me a lot of motion artifacts in the past.

  • When to resize 720x486 for QT web delivery?

    Hello,
    (*correction: disregard deinterlace question)
    I cannot afford to purchase FCP to gain the new compressor engine - so I'm
    re-rendering all my movies (in A.E.) to add a title and have a Copyright visible
    (QT uncompressed).
    This work has to be done on a computer with Mac OS 9 (old software).
    The movies will then be compressed in QT on another computer with QT 7.3.1 using
    the H.264 codec for web delivery.
    QUESTION: When should I resize (from 720x486 to 640x480).
    In Adobe After Effects (OS 9), or over on the newer computer, as I compress with
    the H.264 Codec?
    This might sound like a dumb question - but I'm trying to keep the quality as
    high as possible - and was not sure if resizing one way, or another - might
    incur some artifacts or something.
    Thank you for your help - from those who know.
    Luke Wonderly

    Wow, thanks for more GOOD info!
    OK -
    Yes, I knew about exporting, and QT ability to rescale in the page nicely,
    but I have always been able to tell (on Independent artist - NOT studio videos)
    when its been employed . . . . I need the very best display I can achieve.
    (It also helps differentiate MY Work - from all the rest of the other junk thats
    out there - and why I choose Quicktime.
    These movies will also be the files which the customers will download, save and
    keep on their own computer (to watch and practice with later) - so they need to
    keep the 640 physical size.
    Question:
    Will the DV export codec AFFECT the rgb, yuv color space?
    Movies were originally an older, QT 6.5.2 unsupported Avid Codec - so I had to open in OS 9, and (was cautioned by Ben Waggoner that IF compressing-use the
    motion JPEG B - because it wouldn't effect the color space)
    I export NOW uncompressed (then was for DVD delivery - on limited HD space).
    If I am going to go through opening and re rendering 210 movies (to add titles
    and copyright/logo) would it be the very best way to GO BACK to original (AVID
    Codec) export in DV (highest quality) then import to A.E. - add titles, THEN
    take it over to other computer and export H.264 to .mov?
    This is tough stuff, huh!
    Or, since they are already in uncompressed state - (or was color spaced screwed
    up then - when I exported uncompressed?) should I just import to A.E. do the
    titles - and export to DV.
    What is the HIGHEST quality, and safest way you guys think it should be done?
    Man, looks like I really need some expert knowledge here -
    Would you still be willing to offer any more help, or advice?
    Thanks if you still got the patience!
    Luke

  • H1 footage & premiere deinterlace encoder questions.

    Hi all,
    I recorded some footage on a H1 in 60i HD (HDV). Used Premiere CS3 preset HDV 1080i 30 (60i) to capture it.
    Capture question...
    1. Should i buy cineform aspect or prospect HD to convert it before editing in premiere CS3?
    Adobe media encoder questions...
    2. Should i deinterlace the footage first by selecting "Deinterlace" in the from the output tab?
    I was reading some adobe livedocs info that its a good idea to do so, even though canon says their 60i, 30f, 24f is as good as progressive.
    3. should i choose "HDTV 1080i 29.97 High Quality" or "1440 x1080i 29.97 High Quality"?
    The H1 captured output from tape is 1440x1080i. The encoder HDTV preset outputs 1920x1080, the other preset obviously outputs 1440x1080.

    1. I use Cineform with great results and would indeed convert it to cineform BEFORE editing. It converts the 4:2:0 colorspace to 4:2:2 and is better to edit with, along with the fact it also coverts long GOP to almost a frame based compressin.
    2. Cineforms method of "deinterlaceing" is different than Adobe's.
    I have tried both. It seems that Cineforms is better. There is a LONG post under HDV to SD conversion. Read it.
    But, it seems to me that if you keep it interlaced, motion seems smoother and more natural.
    Adam Wilt did a good review of the HL and stated that the "F" frames are really progressive frames made from interlaced CCD's. The frames do lose a little vertical resolution.
    I would either shoot in F and keep it there, or shoot in I and keep it there.

  • A question about deinterlacing on export in CS4

    I've always found the "deinterlace" checkbox option in Adobe Media Encoder gives substandard results when encoding to WMV (and possibly to other formats, but I'm typically exporting to WMV), particularly when you're exporting at the true frame size. My workflow for fullsize projects which matter has typically been to export a DV AVI, then do the deinterlace/compress stage in Windows Media Encoder which produces distinctly better output.
    Is this now fixed in CS4? Or have I missed a simple trick which solves the problem in CS3.
    Best,
    Keith

    CS4 now decides for you to deinterlace or not based on the export option and source material. Which is why there is no longer a "deinterlace" checkbox in the export options.
    You will have to decide yourself on the quality. You can still perform your workflow if you wish to compare.
    On the flyout menu there is an render high quality option. The main purpose is to improve scaling. It does improve scaling quality but increases render time significantly. I havent tested to see if it also improves deinterlacing where appropriate.

  • To deinterlace or not to deinterlace, is the question

    hi!
    i've imported a movie from my DV camera, and after i edit him i want to convert it to dvd and burn so that i can watch on any dvd player. having this should i deinterlace my movie? it's just that i'm afraid of not interlacing and then i get all those nasty lines (specialy because i will want do put it playing on mac's dvd player) :S, but on the other and if i do not need do deinterlace i do not decrease vertical resolution to my movie, which guarantees better results. what should i do?
    sorry for my bad english...
    cheers!

    you wont need to deinterlace, you'll be fine just outputting to DVD

  • Deinterlace/single field question

    G'day folks
    I am prepping some PAL DV files for computer playback using QTPro's "Export" feature. However most of the actual work will be done on Windows. Due to the video content I want still frames to display just one field, not a blend of the upper and lower fields. Using H.264 and ticking the "deinterlace" button does not work, even if I have previously set "single field" in the source video and saved as self-contained.
    I can get it to work properly using Visual Hub, so the technology is there. But is there a way to get my colleague on a Win box to achieve the desired result?
    TIA

    If you input 1080i, then both fields are simply combined into one frame. In iMovie 08/09 DV had one field dropped leaving a low rez "frame". With 1080i, the odd field is always dominant. With DV, the even field is always dominant.
    I don't do much HD work. But for DV I would use iMovie 06.
    iMovie 08/09 uses 'single field processing' meaning every other horizontal line of the video is thrown out, which reduces the sharpness of the footage. iMovie 06 uses ALL of the image to form the video.
    If your primary workflow is editing DV clips and making DVDs, iMovie '06 is better suited. Your movie will arrive at iDVD in DV format, which is an ideal match for making a DVD: same resolution, same pixels aspect ratio, and original quality. If you share your movie from iMovie '08 / 09, it gets re-rendered at 640x480 or less, and then iDVD upscales it back to 720x480. The end result is obviously not as good.
    iMovie 06 and iDVD 08/09 is a terrific combination.

  • Plug-In Question

    Hi there.
    I'm new to the use of visual effects and had some questions I just couldn't find anywhere... Perhaps it is because they are so basic, so my apologies if these are stupid questions. My projects have usually been on 16mm, but I recently moved to video and was exploring my options on getting the film look.
    Firstly, I was looking into either Twixtor or Magic Bullet for deinterlacing and 24p conversion. Now to use these plug-ins, I first have to transfer my footage into After Effects correct? From what I understand, I have to use After Effects to use either of these plug-ins, and then transfer it back to FCP.
    So at the moment I only have FCS2, but to use either of these programs am I looking at an investment in both After Effects and Twixtor/Magic Bullet?
    Thanks for the help.

    I'm definitely looking to get a Panasonic in the near future. All these programs are mostly for me to get accustomed to before I shoot a documentary this upcoming summer.
    I'm still a little confused, and i think it is because my question is more basic than you think it is.
    When you say that Twixtor/Magic Bullet are AE plug-ins that work in FCP, does that mean owning After Effects is necessary to use them? In other words, I have FCP. In order to use Twixtor do I need After Effects, then Twixtor?
    Thanks for the quick response.

  • Premiere / Media Encoder CS5 and deinterlacing

    Maybe it's just me, but I'm not at all happy with the deinterlacing results from Premiere CS5 and Media Encoder. I've been using Premiere since CS3 and hoped for improvement in CS4, but no. And now with CS5, it's still just not good...
    This is what I do in an nutshell: I edit 1080i HDV video in Premiere (using 1080i timelines) and export it to several formats, including 720p for Youtube. Now, whatever I do, I alway get some staircasing / aliasing on deinterlaced output. Not really bad and hardly noticable on video, but very noticable on logo's (which - ironically - are progressive to start with). Preview in Premiere looks fine, but as soon as I put the content through media encoder: staircasing. Even with the new "Use maximum render quality" option enabled.
    When I output my 1080i timeline as 1080i MPEG2 video everything looks pristine (motion adaptive deinterlacing is then done upon playback by my nVidia graphics card). When I leave all rendering options the same and export 1080p or 720p - leaving deinterlacing to media encoder - staircasting is easily visible on logo's and detail is lost in the video compared to real-time deinterlaced output by my graphics card.
    I guess my main question is: Why can't Adobe get this right? nVidia and ATI can do it very well and in realtime on their graphics cards (and have been able to do so for several years now) and freeware filters such as the ones used by ffdshow do a far better job too.
    Why can't a high-end and expensive video editing solution like CS5 get this right? Hugely disappointing

    Hi Todd, Steve,
    I'm running CS5 a Cuda enabled system  (Intel Core i7 965, 12 GB DDR3, MSI GeForce GTX 285 with 2GB memory and a secondary GeForce 9400GT 1GB, running Windows 7 Ultimate 64 bit).
    Cuda support is enabled in premiere CS5, as my timelines are yellow even when I enable cuda accelerated effects such as color correction. So no problems there.
    I work with 1080i50 HDV and AVCHD files, which I normally use in a HDCAM EX 50i timeline (I use HDCAM EX because that enables me to render all effects and graphics in 1920x1080, instead of a HDV timeline which renders everyting in 1440x1080). I have tried using native HDV timelines too, but that reduces horizontal sharpness on our logo and on sceen graphics, while the deinterlacing problems I encounter are exactly the same.
    The problem is that when I export an HDCAM EX timeline to 720p, I get noticable degradation in image sharpness and visible jaggies on our on-screen logo. The same applies when I export to 1080p. When I export to 1080i (so no deinterlacing is done by Premiere or Media Encoder), the file looks great on playback (deinterlacing done in real time by my nVidia card).
    For the last few weeks I've been working a round the problem in a way which is not ideal for me, but which does seem to provide better results:
    I now work with HDCAM EX 25p timelines instead of 50i, and I set the 'field options' flag for all my clips to 'always deinterlace'. This results in a loss of temporal resolution on my timelines (which I don't really want, since I also export to PAL 50i) but it results in noticably better sharpness and less aliasing on 720p output.
    I'd love to share my on-screen logo title if you wish to have a look at it. Our logo contains a circle which shows easily visible jaggies when deinterlaced incorrectly. Please let me know how to get the files to you.
    Thanks,
    Eric

  • 16:9 JES deinterlacer slow motion

    I am trying to use JES deinterlacer to slow down a clip but cant seem to get the output in 16:9. I have followed the instructions below and keep checking the 16:9 box but the output is always squashed 4:3. I've tried fiddling with the options and cant seem to find an answer.
    Does anyone have any ideas?
    From JES instructions.,..
    Slow-motion and fast-motion: In JES Deinterlacer's Project tab select "Project kind: Standards conversion". Select "Custom" and set "Movie speed: 0.5" for 0.5x slow-motion, 2 for 2x fast-motion and so on for other speeds. You can select speed ratios in the range of 0.1-10x with 0.01 decimal precision (with v2.7.1 the 0.1x may stall and with 0.2x the audio is out of sync. 0.25-10x are OK). THEN set the other options:
    For NTSC use "Settings/Height: 480, Width: 720, Frame rate: 29.97".
    For PAL use "Settings/Height: 576, Width: 720, Frame rate: 25".
    Check "Blend" and let "Scale only" remain unchecked.
    In the Output tab set "Video output/Compressor/Export: DV Stream.../DV Format: DV, Video Format: NTSC/PAL, Scan Mode: Interlaced, Aspect ratio: 4:3/16:9, Audio Format: Unlocked/Locked, Rate: 48.000 kHz".
    Let "Progressive (out)" remain unchecked. (Leave the other options as defaults).
    Select the output folder via "Put..." and click "OK" to start processing.

    I exported the clip as a full quality clip from iMovie5 (had to go back to it after audio trouble importing from my HDR-HC1). I get a .mov file and then use that as an input file for JES. The 'converted' file is a .mov.proj.dv file which looks like a squashed 16:9 in 4:3 window in QT. I tried importing it back into iMovie and it imports as a 4:3 with black side bars - no letterboxing. The project I am working on is a 16:9 HD format.
    I havent tried any other exporting yet as I'm fairly sure if they look that way in QT and imovie then that's probably how they are.
    I do notice that when I make all the settings in JES, as I outlined in my question post, once I change the aspect ratio to 16:9 and then close that window and open it again, the setting seems to have defaulted back to 4:3. I tried it several times... (Ver 2.7.3 - I think that's the latest)
    I really appreciate any advice.

  • Visual Hub questions (related to Apple TV) and others

    Hey guys,
    Just bought Visual Hub to convert my stuff.
    What do you recommend for the settings ?
    Right now this is what I use : (NOTE : I translated on the fly from french so maybe it's a bit different)
    -To : iTunes
    -Optimize for : Apple TV 5.1
    -Quality : BIG QUESTION HERE : I want as close as 1080p as can be, but if the source file I'm -converting is poor quality, won't this make a big file for nothing ? What should I base my choice on?
    -H.264 enabled
    -In advanced, I select double-pass and deinterlace, audio settings = initial
    Nothing else is tweaked in advanced
    so what do you think ?
    I have a couple of other questions :
    - What if I deinterlace a video that is not interlaced in the beginning ?
    - How can I determine if a video is interlaced ?
    - How can I tell if a video is in 5.1 or even full HD ?
    - How can I make sure to get 5.1 on my Apple TV
    - Can Apple TV play 1080p ? or is it just upscaling ?
    - If A TV can output 1080p, where can I find sources file of such quality ?
    Anything else that I should know of related to my questions ?
    Thanks guys !!

    I don't use visualhub so I couldn't suggest settings for you, but.
    but if the source file I'm -converting is poor quality, won't this make a big file for nothing
    Yes, that's all it will do.
    What if I deinterlace a video that is not interlaced in the beginning ?
    You will lose detail
    How can I determine if a video is interlaced ?
    Take a look at it on a progressive scan display such as your mac monitor, you will see horizontal lines in high motion areas when paused.
    How can I tell if a video is in 5.1 or even full HD ?
    See This
    How can I make sure to get 5.1 on my Apple TV
    Your content file need to have an AC3 track and you will need to play the audio through a 5.1 capable receiver or home theatre system.
    Can Apple TV play 1080p ? or is it just upscaling ?
    It can play up to 720p after which it upscales.
    If A TV can output 1080p, where can I find sources file of such quality ?
    More or less nowhere.

  • DVD Question...please help

    This may be a really silly question but... how can I play a regular DVD in quicktime. I downloaded the player in hopes I could watch my DVD's on it but now when I load a DVD nothing happens. Any ideas?
    Thanx

    I'd suggest just using the DVD Player.app or VideoLan.app for DVDs.
    QT Player can open the VOBs in the DVD if the MPEG2 PB component is installed. But the QT Player can open just separate VOBs, it can't play PCM or AC3 audio and it distorts and displays interlacing in PAL so it isn't very good.
    Another alternative is to view the DVD via MPEG Streamclip. It can open the related VOBs all at once and it can play PCM and AC3 audio. It has very handy navigation options via the keyboard or via the GUI. It doesn't currently have an option for deinterlaced playback.

  • Video Levels and Deinterlacing

    I have been trying out Premiere Elements 10 and had a couple of questions. First, if I take AVCHD footage and check the video levels in Premiere Pro, it seems to indicate that the video is 0-255 on the RGB scale. Is there any way to test or see the video levels in Premiere Elements 10? And am I correct in assuming that Adobe converts AVCHD to 0-255, as opposed to Sony Vegas, which does 16-255?
    My second question involves deinterlacing. Is it necessary to select the field options - deinterlace option when rendering web video like WMV, MP4, and FLV, or will Elements do this auotmatically? If I need to select deinterlacing, does it need to be applied on a clip for clip basis, or will selecting it under clip - video options only apply it to video that needs it and not deinterlace, say, a still picture in the timeline that is already progressive? In other words, if selecting deinterlacing is necessary for web videos, does doing so at the clip level cause any quality loss to progressive still images in the project?

    Though PrPro offers many CC Effects, right now, many use AE for Color Grading. However, things might change. Here are some recent threads on what Adobe is planning:
    http://forums.adobe.com/thread/938287?tstart=0
    http://forums.adobe.com/thread/939053?tstart=0
    For the vast majority of my work, I use PrPro, and AE. Remember, PrE is not a "pro-level" NLE, and is basically designed for DVD/BD, and then delivery to sites like YouTube. There are few, if any, broadcast functions in it.
    Good luck,
    Hunt

  • Turning off deinterlace in CS4 before video exporting or a substitute solution

    I have had a problem for a long time now regarding the missing function in the Premiere Pro CS4 Export Settings window when trying to export a video in FLV. When I playback the rendered sequence I get a poor quality video (it also looks like it is stretched vertically). In Premiere Pro CS3 however, I just had to uncheck the Deinterlace option in the Export Settings and then I had a crisp video. Since this option is clearly unavailable in CS4 does anyone know of any subtitution for unchecking the Deinterlace option in CS4? Please note that the output file must be FLV.
    The settings under Clip > Video Options > Field Options... do not seem to do the trick.

    Yes - your question had much to do with the issue.
    I was dealing with very different file formats and even file size.
    I found that m4v is very slow for FCE to process.
    I also found that I can convert files to FCE format much faster in a side program - I used Squared 5 Streamclip to cut the videos down to the clip I required, then saved them as Apple DV/DVCPRO-NTSC with the sound at AAC 48 128. It processed much faster in Streamclip! Then imported without the need to render at upload! I was very pleased to say the least. The output file is .mov and it seems like this makes your FCE much happier!
    We have too many options and not knowing what the options actually mean allows it to work, but really slowly!
    My biggest concern is - why was it sooooooo hard to find information online about this. I searched forever before posting a question. The hardest thing is trying to word the question to be short and find the words that will key into the answer since the words you use is how search works. Search doesn't read vibe - only keywords and frequency creating GOOGLE rank so it will show in the search.
    Thanks for the direction. I had read a lot of things but this was the direction that got me back in control of my MAC.

  • Compressor / Deinterlacing

    Hello -
    I have a few files to encode for the web, and I need some advice on compressor and deinterlacing.
    I exported the video from FCP to Compressor to be encoded @ 640 x 480 with H.624 (2k kbps). I chose the deinterlacing filter, with the sharp algorithm, but the video - when exported - did not look good. I applied the frame controls; choosing out put fields: progressive and deinterlacing: best (motion compensated), and the Compression had 86 hours still to go after leaving the job all night (over 12 hours).
    As a last ditch attempt at getting these jobs to run - with an acceptable end product and an acceptable timeline - before my deadline (Monday), I have applied the deinterlacing filter in FCP, and then submitted the job to Compressor without Compressor applying any deinterlacing.
    Will this provide a deinterlaced final video...?
    I am using FCP 5.1.4 and Compressor 2.3
    I am running the above on a G5 Dual 2GHz w/2 GB DDR2 SDRAM
    My settings for the video in Compressor:
    Name: h26402k
    Description: H.264 for high-bandwidth connections
    File Extension: mov
    Audio Encoder
    AAC, Stereo (L R), 24.000 kHz
    Video Encoder
    Format: QT
    Width: 640
    Height: 480
    Pixel aspect ratio: default
    Crop: None
    Frame rate: 29.97
    Frame Controls: Off
    Codec Type: H.264
    Multi-pass: On, frame reorder: On
    Pixel depth: 24
    Spatial quality: 75
    Min. Spatial quality: 25
    Key frame interval: 24
    Temporal quality: 50
    Min. temporal quality: 25
    Average data rate: 2.048 (Mbps)
    Fast Start: on
    Compressed header
    requires QuickTime 3 Minimum

    Hello All -
    The first scene just finished exporting, and I checked the .mov in Quicktime and the files looks good after applying the deinterlacing in FCP, and not applying any filers or frame controls in Compressor. Thescene took a fraction of the time to export - Around 1.5 hours to export 8 mins of video, compared to 32 hours when using frame control.
    Thanks to anyone who was thinking about answering my question
    Cheers.

Maybe you are looking for