Go Back   Steve's Digicams Forums > Digital Cameras (Point and Shoot) > Hybrid Still/Movie/MP3 Digicams

Reply
 
Thread Tools Search this Thread
Old May 13, 2008, 8:09 PM   #1
Senior Member
 
pgomberg's Avatar
 
Join Date: May 2008
Posts: 110
Default

Deshaker (VirtualDub plugin)

http://www.guthspot.se/video/deshaker.htm


Here is an example of the original, then the processed filed, I did not compensate for the borders, which CAN be done, I just did a quick and dirty example, but the plugin has a lot of options.

Rolling Shutter Compensation

Original file
http://search3.idrive.com/driveway/j...?id=zxrbw82468

Compensated for Rolling Shutter
http://search3.idrive.com/driveway/j...?id=rwooh15412



I open the MOV file with TMPEG, setup the correct AR, resize however I want, yada yada yada, then I save the project file, then CONVERT the project file using "VFAPIConvEN.exe" and then I am able to open the file as a normal AVI in an editing application that can read AVI files. Yes.. it's THAT easy . I have an Aiptek Action HD which I took this video with.
pgomberg is offline   Reply With Quote
Sponsored Links
Old May 13, 2008, 11:56 PM   #2
Senior Member
 
Join Date: Oct 2006
Posts: 584
Default

I use this AVIsynth script to feed a file, eg, FILE0005.MOV, directly to VirtualDub:
DirectShowSource(".\FILE0005.MOV")
# SelectEvery(2,0)
The second line gets rid of every second frame when cameras (like the HD700) put out two identical 30fps frames instead of one 15fps one... Usually it is commented out with that # symbol...

If you have a Direct Show decoder, such as CoreAVC, it will be used to handle the AVC decode, you don't need a VfW codec.


Trevmar is offline   Reply With Quote
Old May 14, 2008, 12:23 AM   #3
Senior Member
 
pgomberg's Avatar
 
Join Date: May 2008
Posts: 110
Default

I actually am using CoreAVC Pro version 1.7.0 and TMPGENC opens it using the Directshow API as the VFAPI Plugin is set to Directshow media file reader. It's a great program. I have used AviSynth in the past though for MOV files.. but for what I'm doing this seems way faster, drag and drop, don't have to edit any script files, and an actual AVI file is created, rather than an AVS file. Avisynth is a great tool however.
pgomberg is offline   Reply With Quote
Old May 14, 2008, 12:55 AM   #4
Senior Member
 
Join Date: Oct 2006
Posts: 584
Default

Er, um, the AVS file just contains the above text, and no huge intermediate file is created. If your aim is to get to use DeShaker in VirtualDub, then all you should want to do is get the MOV file to VirtualDub, which AVIsynth does without needing any intermediate file. And it uses the Direct Show AVC codec.

There is another problem with a TMPG intermediate file - the more often you compress to an intermediate format, the more detail you lose. I used to use TMPG Xpress to do my 24p pulldown, but now I use AVIsynth because it works properly with AVC.

Sorry to introduce conceptual complexity. I had thought that since you were advocating the use of VirtualDub, you might be interested how easily I DeShake all my video...


Trevmar is offline   Reply With Quote
Old May 14, 2008, 1:19 PM   #5
Senior Member
 
pgomberg's Avatar
 
Join Date: May 2008
Posts: 110
Default

Trevmar wrote:
Quote:
Er, um, the AVS file just contains the above text, and no huge intermediate file is created. If your aim is to get to use DeShaker in VirtualDub, then all you should want to do is get the MOV file to VirtualDub, which AVIsynth does without needing any intermediate file. And it uses the Direct Show AVC codec.

There is another problem with a TMPG intermediate file - the more often you compress to an intermediate format, the more detail you lose. I used to use TMPG Xpress to do my 24p pulldown, but now I use AVIsynth because it works properly with AVC.

Sorry to introduce conceptual complexity. I had thought that since you were advocating the use of VirtualDub, you might be interested how easily I DeShake all my video...

actually the intermediate file is not a recompressed file at all, it's a passthrough file kinda like the AVS file, it's not huge at all. it just sends the video from the MOV file, or MP4 file, or WHATEVER file is being passed through, to the application via the VFAPI codec. It's really easy to use and you don't have to learn Avisynth scripting to do it. i have used AVISYNTH in the past, quite a lot actually, so I *am* familiar with doing this with Avisynth. Just seems more compatible this way since not all editing applications can use AVS files.
pgomberg is offline   Reply With Quote
Old May 14, 2008, 2:15 PM   #6
Senior Member
 
Join Date: Oct 2006
Posts: 584
Default

Oops - thanks. I have used TMPEG a lot, but I had no idea it would work as a frameserver. I better take a closer look...

Trevmar is offline   Reply With Quote
Old May 14, 2008, 5:39 PM   #7
Senior Member
 
pgomberg's Avatar
 
Join Date: May 2008
Posts: 110
Default

It's actually the VFAPI Reader Codec that does the frame serving. You can read about it here:

http://www.doom9.org/index.html?/mpg...ader_codec.htm

It can frameserve d2v/tpr/aup files. The only thing is I believe it actually writes the AUDIO output to the file if you tell it to serve the audio also.

"VFAPI acts like a regular video compression codec except that it isn't a codec, but more like a wrapper. A brief explanation on how it works: VFAPI gets uncompressed RGB data from the serving application (DVD2AVI, AviUtl or TMPG) and wraps it into an AVI so that the file can be opened in any application which supports the AVI format."

So, it's pretty cool


I do have a question for you though.. I've not done much DESHAKING.. what is your experience with it and how do you usually render your output files? With or without boarders? Personally I freakin' LOVE VirtualDub and .. all it can do. Thanks.
pgomberg is offline   Reply With Quote
Old May 14, 2008, 9:58 PM   #8
Senior Member
 
Join Date: Oct 2006
Posts: 584
Default

The "without borders" is a misnomer. Actually the advantage of 'borders' is Zoom compensation. Take a look at this FAQ:
http://www.guthspot.se/video/deshaker.htm

I generally set Deshaker up as "Camcorder has rolling shutter" ticked, and default values for Pass 1 (remember that in HD this gives even better accuracy than in SD). Then in Pass 2 I use "Adaptive Zoom (some borders)" (which enables the incredible Zoom compensation), "Use previous and future frames" with +10 and -10 (so I don't lose too many frames at the beginning of the clip) and "Extrapolate colors into borders." I manually re-sync the audio and video in Vegas ('cos I'm a perfectionist, and don't want Virtualdub to recompress the audio).

Deshaker is a killer application - I just don't worry about jerky handheld footage any more - I can relax my camera-holding hand, and concentrate more on the staging as a result...

Trevmar is offline   Reply With Quote
Old May 14, 2008, 10:02 PM   #9
Senior Member
 
pgomberg's Avatar
 
Join Date: May 2008
Posts: 110
Default

You know.. you can just set the audio to direct stream copy and just change the ms sycn options to adjust the audio forward or backward, without recompressing it Gotta love VirtualDub
pgomberg is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 2:46 PM.