| Author |
Message |
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
So, first of all, hi guys! New user. Long time reader, etc. etc. :p Hopefully one of you wizards can lend me a hand here.
Here's my problem. I've got security footage from an art gallery, on VHS tapes. Nearly 60 hours worth. I need to capture this to edit in Final Cut. My only option at this point is to record miniDV (sony camcorder) from the VCR, then capture that in final cut.
Now here's the kicker - it's three or four cameras all recorded onto one tape - simulataneously! So when I watch the footage on a regular VCR, it's an epileptic flicker show - one frame from each camera, in quick succession.
To clarify, in case I'm not explaining myself well - imagine three cameras, A, B and C. So, the tape plays out like this: Single frame from A, single frame from B, single frame from C, ad nauseum. A,B,C,A,B,C,A,B,C. I haven't tried to capure anything yet.
Does anyone have experience dealing with this sort of thing?
Thanks a lot.
|
|
guns1inger Member
Joined: 01 Apr 2004 Location: Miskatonic U
|
|
What you need to do is capture as is, then decimate the clips to remove all the footage that relates to the other cameras. This should be simple because there should be a definite pattern to the way the images are recorded.
I know this can be done relatively easily using avisynth or virtualdub. Not so sure what tools you have available on the Mac side of things.
You would have to repeat the exercise 4 times, with slightly different decimation each time. What you will get are 4 clips, each running at around 7 - 8 frames per second.
The other alternative is to look at the recording unit and se if you can take an output feed in the same way the monitor does, and work from that.
_________________ The views expressed in this post are mine alone, unless plagiarised from others
Read my new blog here
|
|
gadgetguy Contestant
Joined: 14 Feb 2002 Location: Michigan, USA
|
|
Experience? no. But it should be fairly simple to separate them out into 4 videos. For instance in AviSynth, the SelectEvery() filter can be used to select only the frames from one camera and piece them together as a single sequence.
_________________ "Shut up Wesley!"
-- Captain Jean-Luc Picard
Buy My Book
|
|
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
Thanks for the help, guys. Will AVI synth compensate for a drop-frame timecode, if that issue should arise?
I was hoping to avoid capturing all of the footage, so a pre-capture solution would be preferable - but the info is invaluable, guys. Cheers.
|
|
jagabo Member
Joined: 09 Dec 2005 Location: none
|
|
You probably won't be able to fix this during capture with a DV camcorder. I don't know Final Cut but if it has an option to speed up the video you can try setting it to 3x. Ie convert from 29.97 fps to 89.91 fps. Most programs that do this will simply throw away two out of every three frames in a repeatable pattern:
A B C A B C A B C --> A - - A - - A - -
If that works, trim one frame off the start of the video and repeat:
B C A B C A B C --> B - - B - - B - -
Then trim another frame off the start:
C A B C A B C --> C - - C - - C - -
If drop frame or something else causes a phase change
A B C A B A B C A B C --> A - - A - - B - - B - -
You can just go back and cut/paste from the three files you created (hopefully there won't be too many discontinuities).
|
|
2Bdecided Member
Joined: 21 Nov 2007 Location: United Kingdom
|
|
Check that it really is individual frames (pairs of fields) - if it's individual fields then capturing via DV could be a bad idea - it'll drop back to field compression mode for the entire capture, and on some devices this dramatically reduces quality.
Cheers,
David.
|
|
edDV Member
Joined: 06 Mar 2004 Location: Northern California, USA
|
|
Check out the manufacturer of the security VCR. They provide solutions to extract and pad out playback from each camera channel.
_________________ Recommends: Kiva.org - Loans that change lives.
http://www.kiva.org/about
|
|
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
2B, great point. Thank you.
And thanks ed.
|
|
A_51 Member
Joined: 16 Oct 2005 Location: Here & there
|
|
Why would you want to edit security cameras footage? Ones you manipulate the footage is not admissible on court.
Nevertheless what you are encountering is the result of the security camera footage being multiplexed in a time laps recorder. The topology goes like this. The cameras are connected to a multiplexer device, this device arrange the video coming from the cameras on a matrix fashion on a monitor for easy viewing and transmit the video to a recorder on the way you describe for archiving purposes. To play back the footage the signal needs to be feed back to a multiplexer for re-arranging.
Most probably the cheapest way to deal with what you want to do is to prepare a patch cable. With BNC connector on one end and regular RCA in the other and capture the output from the playing monitor with a portable recording device.
_________________ I want to believe....
|
|
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
Thanks, A_51
The security footage is for documenting purposes. I work freelance for advertising, and one of the things they like to do is prepare short videos that demonstrate how practical or stunt campaigns take place, for awards purposes.
...it is rater ugly footage anyhow. But thanks for the help. I think I'll be able to access the recording device again. They're going to get back to me with the specifics, so if edDV was barking up the right tree, I may find an even simpler method. Here's to hoping!
|
|
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
...alright. So I'm going to do a little research on my own, but here's the technology, incase any of you have further hints or tricks:
Duplex Digital Video Multiplexer from Deltavision, model M4000E
and a Sony Timelapse videocassette recorder, model SVT-LC300.
|
|
A_51 Member
Joined: 16 Oct 2005 Location: Here & there
|
|
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
oldandinthe way Dissenter
Joined: 14 Mar 2006 Location: With the other crabapples
|
|
I'd buy a Magnovox or Funai DVDrecorder at Walmart for $50 to do the capture.
Reading from the DVD is less prone to problems and faster than capture from the camera. If the job doesn't bring in enough to allow this investment you need to reconsider your activities.
|
|
jagabo Member
Joined: 09 Dec 2005 Location: none
|
|
| oldandinthe way wrote: |
I'd buy a Magnovox or Funai DVDrecorder at Walmart for $50 to do the capture.
Reading from the DVD is less prone to problems and faster than capture from the camera. If the job doesn't bring in enough to allow this investment you need to reconsider your activities. |
Video from three separate cameras is multiplexed frame by frame is going to kill MPEG compression.
|
|
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
Yeah, I think it'd be pretty gross. Thanks for the tip though, oaitw.
|
|
Cornucopia Patently Pending
Joined: 22 Oct 2001 Location: E-Cnt. IL, USA (AGAIN!)
|
|
| A_51 wrote: |
Why would you want to edit security cameras footage? Ones you manipulate the footage is not admissible on court.
Nevertheless what you are encountering is the result of the security camera footage being multiplexed in a time laps recorder. The topology goes like this. The cameras are connected to a multiplexer device, this device arrange the video coming from the cameras on a matrix fashion on a monitor for easy viewing and transmit the video to a recorder on the way you describe for archiving purposes. To play back the footage the signal needs to be feed back to a multiplexer for re-arranging.
Most probably the cheapest way to deal with what you want to do is to prepare a patch cable. With BNC connector on one end and regular RCA in the other and capture the output from the playing monitor with a portable recording device. |
What you just said is patently false.
I've done a number of these before for security, police and governmental clients. The ORIGINAL is ALWAYS admissible in court, and the "Manipulated" (what a misuse of a word!) version is often admissible as a "simplified rendition" for demonstration purposes. One must still follow chain-of-evidence rules and submit affadavits about the exact processes used (+their need/significance, their pros/cons and their resulting effect), but that's no biggie. (**Note: you may be called to testify about the veracity of results, and your credentials)
Also, FREE is certainly cheaper than the way you suggested, and AVISynth/Virtualdub is free.
Scott
_________________ "You don't know what you got, until you lose it".--John Lennon
|
|
A_51 Member
Joined: 16 Oct 2005 Location: Here & there
|
|
Hello again.
Cornucopia you just said it. In your experience you have made your self liable and had to proof that the process is clear and had follow the proper procedures in order to be acceptable. The fact that our friend didn't know about the nature of the multiplexing of the video shows the lack of deep understanding of security CCTV Equipment and use. That could land him on deep waters if proper procedure is not follow. I don't think is anybody intention to propagate misinformation. What might happen is misinterpretation.
That being said, he stated that his purpose is the creation of training material for his client. The easiest way of doing to me is letting the equipment do what its design for, there is no need to re invent the wheel. Just capture the processed output from the equipment and edit the footage as needed. Simple....
_________________ I want to believe....
|
|
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
...sadly, I have no access to original equipment.
|
|
RiverOtter Member
Joined: 03 Dec 2007 Location: Canada
|
|
....and things get sticky. I decided to directly capture a tape, to see exactly what I'm dealing with. It's field by field.... and almost...random?
I broke a chunk off to see if any of you dudes can help me out.
http://www.mediafire.com/?5f5rlbvg63g
Cheers, fellas.
|
|
gadgetguy Contestant
Joined: 14 Feb 2002 Location: Michigan, USA
|
|
I had some trouble seeking through the mov format so I couldn't get a working AviSynth script, but a frame by frame analysis shows a definite pattern. 121212343434121212343434 So it should be possible to extract them automatically.
_________________ "Shut up Wesley!"
-- Captain Jean-Luc Picard
Buy My Book
|
|
jagabo Member
Joined: 09 Dec 2005 Location: none
|
|
Yes, the frames contains fields from different frames. There is a repeating pattern every 48 frames. I converted the video to uncompressed RGB with the latest VirtualDub then used this AVISynth script:
#ASYNTHER AVISource
AVISource("C:\Documents and Settings\Johnny\Desktop\uncomp.avi")
trim(22, 0)
SelectEvery(48, 0, 4, 8, 22, 26, 30)
Bob()
SelectEvery(2, 1)
LanczosResize(320, 240)
to make this:
sample1.avi
and
#ASYNTHER AVISource
AVISource("C:\Documents and Settings\Johnny\Desktop\uncomp.avi")
trim(22, 0)
SelectEvery(48, 0, 4, 8, 22, 26, 30)
Bob()
SelectEvery(2, 0)
LanczosResize(320, 240)
to make this (the opposite field):
sample2.avi
You can work out the rest.
Ah, but I missed some frames. In the gap between 8 and 22 there frames from the same camera but the opposite field. I'll post an updated script later.
OK, here's all unique fields (BOB'd into frames) from one camera, starting at frame 22 (field 44, because it was easiest to count from there).
#ASYNTHER AVISource
AVISource("C:\Documents and Settings\Johnny\Desktop\uncomp.avi").trim(22, 0)
AssumeTFF()
Bob()
SelectEvery(96, 0, 8, 16, 25, 33, 44, 52, 69, 77, 85)
sample3.avi
|
|
mschult Member
Joined: 08 Dec 2007 Location: United States
|
|
I have a slightly different issue. I have security cameras and I have purchased easy media creator 10 suite. I have taped my security footage on vhs. I have copied it to my DVD using a DVD recorder. It created .vob and .ifo files. I have tried converting to every format I can to play this back on one of my other dvd players not on my computer. I keep getting errors that say disc error on the players. I have checked your site and found the correct dvd's to use for recording and also what format works for each. They still get the errors. I am frustrated and don't know what to do. Am I missing something here on the conversion and recording.
I have footage of some people that came to my house to do criminal damages for the 5th time in a week. I really want to take this into court and have something where I can see these people's faces better. If I can't convert it to something that plays on any dvd player then I am limited to the vhs tape which is ugly to view. Looks better on a DVD on my DVD recorder with a larger screen tv.
I am purchasing a new DVR card that will split each of my cameras into its own channel so that will help future recording. Just waiting on delivery now.
|
|
jagabo Member
Joined: 09 Dec 2005 Location: none
|
|
| mschult wrote: |
| I have taped my security footage on vhs. I have copied it to my DVD using a DVD recorder. It created .vob and .ifo files. I have tried converting to every format I can to play this back on one of my other dvd players not on my computer. I keep getting errors that say disc error on the players. |
Did you finalize the discs? Until they're finalized there is no master table of contents. Regular DVD players need that.
|
|
mschult Member
Joined: 08 Dec 2007 Location: United States
|
|
I did do the finalize on the dvd recorder. It still won't play on any of my dvd players. It will however play on my computer but that does not help me in court.
|
|
A_51 Member
Joined: 16 Oct 2005 Location: Here & there
|
|
Make sure that your video and content structure is withing the specs. This have been taken from this same site.
Technical Info for DVD-Video
PAL
Video:
Up to 9.8 Mbps* (9800 kbps*) MPEG2 video
Up to 1.856 Mbps (1856 kbps) MPEG1 video
720 x 576 pixels MPEG2 (Called Full-D1)
704 x 576 pixels MPEG2
352 x 576 pixels MPEG2 (Called Half-D1, same as the CVD Standard)
352 x 288 pixels MPEG2
352 x 288 pixels MPEG1 (Same as the VCD Standard)
25 fps*
16:9 Anamorphic (only supported by 720x576)
Audio:
48000 Hz
32 - 1536 kbps
Up to 8 audio tracks containing Dolby Digital, DTS, PCM(uncompressed audio), MPEG-1 Layer2. One audio track must have MPEG-1, DD or PCM Audio.
Extras:
Motion menus, still pictures, up to 32 selectable subtitles, seamless branching for multiple storylines, 9 camera angles. And also additional DVD-ROM / data files that only can be read by computer DVD drives.
Total:
Total bitrate including video, audio and subs can be max 10.08 Mbps (10080 kbps)
* Mbps = million bits per second
* kbps = thousand bits per second
* fps = frames per second
For more technical DVD-Video details read the DVDDemystified DVD FAQ section 3.4
NTSC (NTSC Film)
Video:
Up to 9.8 Mbps* (9800 kbps*) MPEG2 video
Up to 1.856 Mbps (1856 kbps) MPEG1 video
720 x 480 pixels MPEG2 (Called Full-D1)
704 x 480 pixels MPEG2
352 x 480 pixels MPEG2 (Called Half-D1, same as the CVD Standard)
352 x 240 pixels MPEG2
352 x 240 pixels MPEG1 (Same as the VCD Standard)
29,97 fps*
23,976 fps with 3:2 pulldown = 29,97 playback fps (NTSC Film, this is only supported by MPEG2 video)
16:9 Anamorphic (only supported by 720x480)
Audio:
48000 Hz
32 - 1536 kbps
Up to 8 audio tracks containing DD (Dolby Digital/AC3), DTS, PCM(uncompressed audio), MPEG-1 Layer2. One audio track must have DD or PCM Audio.
Extras:
Motion menus, still pictures, up to 32 selectable subtitles, seamless branching for multiple storylines, 9 camera angles. And also additional DVD-ROM / data files that only can be read by computer DVD drives.
Total:
Total bitrate including video, audio and subs can be max 10.08 Mbps (10080 kbps)
* Mbps = million bits per second
* kbps = thousand bits per second
* fps = frames per second
For more technical DVD-Video details read the DVDDemystified DVD FAQ section 3.4
DVD File/Folder Structure
Explanation:
. BUP = Backup files of the IFO files.
. IFO = The IFO files includes information such as chapters, subtitle tracks and audio tracks.
. VOB = The VOB files contains the actual video,audio,subtitles and menus.
Folder Files Explanation
AUDIO_TS (undefined) DVD Audio
VIDEO_TS VIDEO_TS.BUP
VIDEO_TS.IFO
The first video play item, IFO, usally a copyright notice or a menu
VIDEO_TS.VOB The first video play item, VOB
VTS_01_0.BUP
VTS_01_0.IFO Title 01, IFO, usually the main movie
VTS_01_0.VOB Title 01, VOB 0, the menu for this title
VTS_01_1.VOB Title 01, VOB 1, the video for this title
VTS_01_2.VOB Title 01, VOB 2, if larger than 1 GB it will be splitted into several vobs
VTS_01_3.VOB Title 01, VOB 3
VTS_01_4.VOB Title 01, VOB 4, up to 10(0-9) VOB files if necassary
VTS_02_0.BUP
VTS_02_0.IFO Title 02, IFO, usually movie extras
VTS_02_0.VOB Title 02, VOB 0, the menu for this title
VTS_02_1.VOB Title 02, VOB 1, the video for this title
VTS_xx_x.BUP
VTS_xx_x.IFO And so on
VTS_xx_x.VOB
VTS_xx_x.VOB
VTS_99_9.VOB Up to 99(1-99) titles with max 10(0-9) VOB files each
_________________ I want to believe....
|
|
|
|