2 Replies Latest reply on Sep 5, 2010 9:11 AM by the_wine_snob

    Breaking .vob files into taggable, searchable clips

    RKCKRC

      Hi-

       

      After hours of searching and experimenting, I am embarrassed to bring what should be an incredibly simple question to this group (I am expert in PS/LR and new to PE).

       

      I am helping my father archive and edit 40 DVD's video transfers of old 8mm films. Each DVD has approximately 5-6 distinct "clips" (not in the PE sense) of unique subject matter (e.g. beach vs. christmas).

       

      My initial goals are simple:

       

      1. Import all the DVD's into a common source catalog.

      2. Break each DVD (.vob file) into its distinct clips and date and tag each clip to allow for search

      3. Initial "clean-up" edit of each clip to set appropriate in/out points and to edit out excess and unwanted video

       

      The result of this first pass would be a a catalog (not literal in the PE sense) of the approximately 200 clips from the full collection, initally edited and tagged for easy searching.

       

      At a later time, I expect this clip library would serve as the source for the creation of multiple movies.

       

      Frankly, I am stumped:

       

      1. The only technique that seems to come close is to identify each clip, set in/outs and elimate unnecessary content, and create a new file using the "work area bar" to Share/File technique.

       

      2. I can turn each .vob into distinct, named clips using the standard techniques -- but I can't figure out how to tag these clips or populate the Organizer database with them for future search. In my naivete, the organizer only seems to recognize files rather than clips.

       

      I would appreciate any guidance how to reach my goals, either a way to tag/access clips, a better way to create files -- or hopefully a much simpler technique that I'm missing.

       

      Thanks in advance for the help.

       

      Ken

        • 1. Re: Breaking .vob files into taggable, searchable clips
          nealeh Level 5

          Scenalyzer gets good recommendations in these forums although when I trialled it I found it created too many scenes. In view of the distinct nature of your clips it should not be too onerous to go through the clips on the timeline pressing [Ctrl]+[K] to split the footage. My personal workflow when I need to do something like this is to (a) Import the video to DV-AVI (b) cut the scenes in the budget software 'Magix Movies on DVD (c) Batch export from MoD to DV-AVI Type 2, creating a file for each segment (d) import these to the PRE Organiser. Because MoD performs 'smart rendering' there is no quality loss.

           

          On a general note I do suggest that you don't manage your libraries in .vob format ( a wrapper for MPEG). MPEG is highly compressed and any future work from PRE will recompress them during render. You would be much better to convert these files to DV-AVI Type 2 and use those in your libraries (DV-AVI is only mildly compressed). I have seen reference to the Lagarith lossless codec but I have not used it myself. For conversion to DV-AVI take a look at What tools can I use to convert my video to DV-AVI?

           

          Cheers,
          --
          Neale
          Insanity is hereditary, you get it from your children

          • 2. Re: Breaking .vob files into taggable, searchable clips
            the_wine_snob Level 9

            Though working from VHS tapes and doing the digitizing myself, instead of working from a DVD, I will use a spreadsheet that I created to give me a shot log.

             

            I capture each VHS tape as one file, and then in the NLE program, will Cut that file into pieces, and I will add a Title with necessary info, like subject, date/time, etc. This also will give me a visual point, that is easy to find. I have found that a green, about the color of a greenscreen, with contrasting type color is easy to locate visually, as one shuttles through the file. I also have my spreadsheet, which was partially filled in, during the capture process. In my case, I just Export as either a DV-AVI Type II, or with the Lagarith Lossless CODEC, for later use. Note: these will be large files, but either will provide good material to work with, with the DV-AVI being only slightly compressed, and the Lagarith being slightly compressed, but completely lossless. An alternate workflow would be to create the same "scenes" but to Export with the WAB (Work Area Bar) into discrete files for later use. With my spreadsheet, and the bright green Titles, retrieval is really quite easy.

             

            I recently completed such a Project for a client with 17 VHS tapes, covering about 25 years of his family. I can still grab a segment with my spreadsheet in hand.

             

            In your case, you already have the digitization done, so you will not be sitting for hours, making notes, as I did, and then filling in more details, when I made my Cuts and Titles. You are also starting with heavily-compressed MPEG-2 footage already. In my cited Project, the client had had DVD's created already, but they were done directly from a VHS-DVD deck, and the color and density was not improved in any way. I also did not want to work from the compressed MPEG-2 material, as I knew that the final output would be back to edited DVD's, resulting in a second compression. Instead, I did the digitizing, and told the client to just keep those DVD's as an archive. Had they not been done already, I would have done the same thing, but only for archival purposes. One level of MPEG-2 compression, going to the output DVD is not bad - two MPEG-2 compressions is not a pretty thing, in my book, and I try to avoid it if at all possible.

             

            Though I use ScenAlyzer Live for my Capture, I also find that its Scene Detect can yield too many files, Often this is because of a change in lighting, or because the camera was stopped, and moved, though the real "scene" has not changed THAT much. The algorithms used to determine scene change do not include any AI, so only the editor can make a determination if the scene has really changed. For instance, Liz's 12th Birthday might have a half-dozen "scenes" in it. I would not want these to end up being separate Clips, but grouped into one, for use later on. One has the option of either cutting things apart, or having to assemble them back into a cohesive Clip later on. For me, I'd rather do the cutting, than the assembly, but that is just my mindset.

             

            Good luck with the Project, and I urge you to make as many notes, as you can. Those will prove invaluable later, regardless of which workflow you choose.

             

            Hunt