VideoHelp Forum
+ Reply to Thread
Results 1 to 9 of 9
Thread
  1. Member
    Join Date
    Dec 2019
    Location
    Denmark
    Search PM
    Hey everyone

    First of all I am sorry if this is the wrong forum for the following questions, but going through several threads, it seems like you guys are very competent at getting the best from old video content.

    I was lucky enough to have a grandpa that had a decent videorecorder and the time to document large portions of my childhood. It is these tapes I would like to care for and manage in the best way. Luckily my dad had all of the old PAL VHS and Video8 tapes digitized and stored on several DVDs, so I don't have to struggle with that task. They are therefore all currently in a MPEG 4:2:0 format in a .VOB container. I would like to move them to my cloud, so I can share them with the whole family, but how do I go around this task the smartest?

    My prioritized list:
    1) Split the long tapes into the clips they are made of
    2) Organize the clips with tags (metadata?)
    3) Improve exposure and colour
    4) Improve video quality (Upscale?)

    Which leads me to the following questions:
    - Should I convert the video to another format and compression?
    - Is deinterlacing necessary? And if yes, what and how .
    - What software is easiest (and fastest) to split video with?
    - Are there any "quick" fix to improve video quality in bulk?

    I have attached a sample of the video in question.

    BTW, I don't mind spending some money on software if needed and I am all open ears to any suggestions.
    Image Attached Files
    Quote Quote  
  2. Originally Posted by thjuu View Post
    1) Split the long tapes into the clips they are made of ... What software is easiest (and fastest) to split video with?
    VideoRedo lets you do this without having to re-encode. Based on the clip you provided, I don't think you'll be able to improve it much. Therefore I recommend that you simply cut it, without re-encoding, thus preserving what little quality you have.

    Originally Posted by thjuu View Post
    2) Organize the clips with tags (metadata?)
    I don't know if you gain much putting these tags inside the video files. I would instead do it simply by choosing file names that include those tags (e.g., "Joe's first date; 1985; Los Angeles ... ", etc.). You can also get a simple cataloging program that will do this as well.


    Originally Posted by thjuu View Post
    3) Improve exposure and colour.
    You might be able to make a little improvement here, but those shadows look pretty dense, so I am not too hopeful that is will be worth the effort.

    Originally Posted by thjuu View Post
    4) Improve video quality (Upscale?)
    Upscaling does not improve quality. At all. Even a little. The only reason to upscale is to match other footage when you are intercutting, and even then, many NLEs handle that for you.

    Originally Posted by thjuu View Post
    Should I convert the video to another format and compression?
    It is true that you could make the files a little smaller, but that is all you will accomplish. What you will also do -- and there is no way around this -- is you will degrade the video further. You ALWAYS want to avoid re-compression, and only do it when the improvement made (like noise reduction) is going to more than offset any new artifacts that you will get from the new compression.

    Originally Posted by thjuu View Post
    Is deinterlacing necessary? And if yes, what and how
    Absolutely not. Deinterlacing ALWAYS degrades the image. It can't be helped. There is a lot of mythology that has built up around deinterlacing, and many people still think that it is a good thing. It is not a bad thing, but since it does degrade the image and since, to do it well, it takes quite a bit of computer time, I always advise against it.

    Modern TVs all have really good deinterlacers built in and they do almost as good a job as even the most sophisticated deinterlacing algorithm you can find. This wasn't always the case, so 10-15 years ago lots of people recommended deinterlacing in order to make the video look better when displayed on the then-new HD display. Perhaps, back then, deinterlacing was a good recommendation. Not any more. The only time you need to deinterlace is when you are doing scaling and when you do temporal processing (i.e., using more than one field of interlaced video when processing that video).
    Last edited by johnmeyer; 6th Dec 2019 at 22:37. Reason: clarity
    Quote Quote  
  3. Member
    Join Date
    Jul 2007
    Location
    United States
    Search Comp PM
    +1 to everything johnmeyer stated.

    In addition, your videos are encoded as mpeg-2 which doesn't allow easy and accurate editing and the quality (even with VHS) is not optimal. I'll let others walk you through the best process for editing.
    Quote Quote  
  4. Member
    Join Date
    Dec 2019
    Location
    Denmark
    Search PM
    Thank you for your response. Going into this project I thought that the potential for improvement was better the worse the quality, but I hear you. I will leave the footage untouched for now and limit my effort to clipping and organizing.

    Originally Posted by thjuu View Post
    2) Organize the clips with tags (metadata?)
    I don't know if you gain much putting these tags inside the video files. I would instead do it simply by choosing file names that include those tags (e.g., "Joe's first date; 1985; Los Angeles ... ", etc.). You can also get a simple cataloging program that will do this as well.
    The idea behind putting the information into meta was to make them independent of program and platform, thus avoiding being stuck to the same program forever or having to redo the tagging. Furthermore, I would like to tag people in the video so it easy to find all the footage of my sister for example. I agree that the naming should still contain some of the info by following: "Date_Description".
    Quote Quote  
  5. Originally Posted by thjuu View Post
    The idea behind putting the information into meta was to make them independent of program and platform, thus avoiding being stuck to the same program forever or having to redo the tagging. Furthermore, I would like to tag people in the video so it easy to find all the footage of my sister for example. I agree that the naming should still contain some of the info by following: "Date_Description".
    Most file systems (Mac and Windows for sure) let you have pretty long file names, so in that sense, putting the data in the file name is platform independent.

    You may be able to find a multiplexer that will put meta data into an MPEG-2 file, although I'm not sure meta data was ever part of that spec. However, if platform independence is one of your goals, you have to decide whether you'll be able to find a program, on every possible platform, that can read metadata inside a file. My experience is that metadata reading is hit or miss. As one example, even with something as old and as universal as JPEG (still photos, not video) the EXIF (metadata) information is often only partially available within some programs. What's more, finding programs that can search on the information may take some effort. By contrast, the file name is easily searchable in any operating system, using hundreds of programs and utilities, including the operating system itself.

    I have tagged over 100,000 photo files over the past fifteen years. To reduce the time this takes, I use people's initials, at least for those who appear frequently (e.g., family members).
    Quote Quote  
  6. Video Restorer lordsmurf's Avatar
    Join Date
    Jun 2003
    Location
    dFAQ.us/lordsmurf
    Search Comp PM
    Ugh, no, never deinterlace. It loses quality, sometimes 50%+. Only deinterlace when the delivery demands it, such as Youtube. But your master is still kept as interlaced.

    Never upscale. You gain nothing. There are only downsides.

    If you used a DVD recorder, then Womble or VideoReDo work well.

    No, no quick fix for restoration. It needs time, testing, and usually Avisynth. But DVD source is always compressed/degraded, lossless was what was really needed for quality restoration work.

    For now, you can just sort/clip/etc. And later, you can capture with better quality, and restore/filter those as needed.

    I don't like tagging, prefer to sort by folder, event, date, make notes as needed.
    Want my help? Ask here! (not via PM!)
    FAQs: Best Blank DiscsBest TBCsBest VCRs for captureRestore VHS
    Quote Quote  
  7. Member
    Join Date
    Jul 2007
    Location
    United States
    Search Comp PM
    Originally Posted by johnmeyer View Post
    Originally Posted by thjuu View Post
    The idea behind putting the information into meta was to make them independent of program and platform, thus avoiding being stuck to the same program forever or having to redo the tagging. Furthermore, I would like to tag people in the video so it easy to find all the footage of my sister for example. I agree that the naming should still contain some of the info by following: "Date_Description".
    Most file systems (Mac and Windows for sure) let you have pretty long file names, so in that sense, putting the data in the file name is platform independent.

    You may be able to find a multiplexer that will put meta data into an MPEG-2 file, although I'm not sure meta data was ever part of that spec. However, if platform independence is one of your goals, you have to decide whether you'll be able to find a program, on every possible platform, that can read metadata inside a file. My experience is that metadata reading is hit or miss. As one example, even with something as old and as universal as JPEG (still photos, not video) the EXIF (metadata) information is often only partially available within some programs. What's more, finding programs that can search on the information may take some effort. By contrast, the file name is easily searchable in any operating system, using hundreds of programs and utilities, including the operating system itself.

    I have tagged over 100,000 photo files over the past fifteen years. To reduce the time this takes, I use people's initials, at least for those who appear frequently (e.g., family members).
    There was a member, Spectateswamp (banned) who advocated putting as much info in filename as possible, including geographic location info. He also advocated shooting in short spurts and never editing the videos. You can can guess my opinion on his advice.

    Windows file name length is limited to 255 characters. A lot, but possibly not enough if you add in multiple descriptive tags. Also some programs truncate the file name after about 100 characters or can't read it file until you cut the file length down. I don't know about the new Mac OS(s), but some older Mac OS had issues recognize filenames after a certain length.

    AFAIK and supported by this article, metadats isn't supported in MPEG2

    "Self-documentation
    Technical (coding) information is contained in the MPEG-2 bitstream in macroblock headers, slice headers, picture headers, Group of Picture (GOP) headers, and sequence headers. The sequence header provides data needed before decoding can begin, such as the size of the picture and the frame rate. Because it is so important, the sequence header is usually repeated as often as twice a second.

    The lack of metadata of the type called bibliographic by librarians motivated the MPEG group to develop MPEG-7, a separately standardized structure for metadata to support discovery and other purposes."

    I don't think the OP came back to the thread, but I mentioned this as an issue in a thread where the OP asked about pulling metadata from episodic DVD rips.
    Quote Quote  
  8. "They are therefore all currently in a MPEG 4:2:0 format in a .VOB container. I would like to move them to my cloud, so I can share them with the whole family, but how do I go around this task the smartest?

    My prioritized list:
    1) Split the long tapes into the clips they are made of
    2) Organize the clips with tags (metadata?)
    3) Improve exposure and colour
    4) Improve video quality (Upscale?)

    Which leads me to the following questions:
    - Should I convert the video to another format and compression?
    - Is deinterlacing necessary? And if yes, what and how .
    - What software is easiest (and fastest) to split video with?
    - Are there any "quick" fix to improve video quality in bulk?"

    1. Realistically, nobody will be able to easily play VOB in the cloud, if you're talking about today's casual PC and Smartphone users.
    MP4 is currently the most "universal" format that plays in everything from phone to tv to pc.
    Also, if you don't upload compatible MP4, almost all cloud services will auto-convert, so it's best to do your own conversion (eg. Vimeo, YouTube...).

    2. You will convert, but smart to upload an original COPY of the VOB/MPEG-2 to the cloud as well.
    This way, anyone who is interested in the future to go back and edit from the source, they'll have it.

    3. Deinterlacing
    Yes, smart to IMO because everything MP4 online will generally be progressive.
    Interlacing was good in the old days where analog TV broadcasts didn't have the bandwidth to handle progressive, but no reason today.
    But, you can test and find what works best for you - I generally deinterlace when outputting MP4 edits.

    Usually, your video editing software will do this for you well.
    eg. If you're editing in Premiere, Vegas Video, Avid, Davinci, etc, simply render/output as progressive MP4 and the software usually does the deinterlacing for you.

    4. MPG2CUT2
    Handles DVD MPEG-2 cutting and splicing very well and fast with no modifications to the parts it doesn't touch.

    5. Depends on what you're trying to "improve" - resolution/details, colors, etc.
    Most video editing software can easily copy the changes you've made to one video and apply it to any other by pasting.
    If you're increasing the resolution, don't go with traditional resizing - instead, try AI resizing https://www.cinema5d.com/topaz-gigapixel-ai-video-resolution-upscale/ which has improved greatly in the past decade to actually do some good.

    6. Metadata
    You can add it to any MP4 easily by bringing up Properties -> Details and edit away.

    otherwise, many video editors have a logcial file naming pattern.
    eg. YY-MM-DD - Bill and John at Disney for Birthday

    Do both since the cloud won't usually support MP4 metadata searches, but will on filenames.

    7. Exposure and color
    Best way here is to eyeball it by opening a second window that has a photo or video that is excellent in exposure and color.
    eg. https://www.adorama.com/alc/alc_images/article8394_1.jpg

    Otherwise, you'd need a video editor with a vectorscope/histogram and learn how to edit to that.
    ..........

    Anyways, while you can do all this using open source, once you do a lot of "fixing" - color correction, special effects, transitions, etc - you'll generally want to pick a good video editor (Premiere, DaVinci, Avid, Vegas Video, etc). They do have trials, some are free (Avid Media Composer First and Blackmagic Davinci eg.), and some have "home" versions (eg. Vegas Movie Studio, Adobe Premiere Elements).

    There's a slew of 2nd tier video editing software below the top tier from Cyberlink, Corel, etc - while these will generally work fine, they have far less support and compatibility. (eg. Try to take a Corel video project and give it to someone else to look at and .... they probably can't because they'd have to buy the program. On the flip side, many already have Adobe Premiere (as part of Adobe CC Cloud), etc.) But as long as it works for you, go for it! These programs have been around for years and years because they generally work.

    Many of the home versions have features home users want - auto-collages, stories, nice titles, etc.
    eg. read https://www.adobe.com/products/premiere-elements.html

    .......
    Quote Quote  
  9. Also, if you don't upload compatible MP4, almost all cloud services will auto-convert, so it's best to do your own conversion (eg. Vimeo, YouTube...).
    YouTube at least will auto-compress no matter what, even if a video is uploaded in MP4 with parameters matching those currently used on YT.

    eg. If you're editing in Premiere, Vegas Video, Avid, Davinci, etc, simply render/output as progressive MP4 and the software usually does the deinterlacing for you.
    But how does it compare, against a QTGMC deinterlacing ?

    4. MPG2CUT2
    Handles DVD MPEG-2 cutting and splicing very well and fast with no modifications to the parts it doesn't touch.
    DVBCut is very good too for that purpose. Avidemux can do lossless cutting as well.
    Quote Quote  



Similar Threads

Visit our sponsor! Try DVDFab and backup Blu-rays!