[Cin] Issues related to Motion Plugin

Miroslav Rovis miro.rovis at croatiafidelis.hr
Sun Sep 25 18:47:44 CEST 2022


On 220925-15:40+0200, Miroslav Rovis wrote:
> On 220925-20:08+0700, Georgy Salnikov wrote:
> > On Sun, 25 Sep 2022, Miroslav Rovis wrote:
> > 
> > > > If the user needs only a smoothed video, perhaps he
> > > > should not do it with the Motion plugin. The ffmpeg's
> > > > vid.stab plugin is much easier to use, it is faster, and
> > > > gives very good results.
[...]
> https://ffmpeg.org/ffmpeg-all.html#vidstabdetect-1
> https://ffmpeg.org/ffmpeg-all.html#vidstabtransform-1
[...]
So I thought the best way to stabilize a video was...
> considered that I know how to stabilize a video, as in:
> https://lists.cinelerra-gg.org/pipermail/cin/2022-September/005565.html
> The slow way, the Motion Plugin way.

But now, thanks to:
> > Georgy Salnikov
> > NMR Group
> > Novosibirsk Institute of Organic Chemistry
> > Lavrentjeva, 9, 630090 Novosibirsk, Russia
> > Phone   +7-383-3307864
> > Email   sge at nmr.nioch.nsc.ru
> > _______________________________________________________________________________
[...]
> Thank you again!

It was in essence just one line:
NUM=40 # or NUM=31 for 70% larger file w/o noticeable
	   # improvement
$ ffmpeg -i VID_20220910_133415_5.mp4 -c:v libvpx-vp9 -crf $NUM \
	-b 0 -vf vidstabdetect=shakiness=10:accuracy=15 -pass 1 \
	-row-mt 1 -f null /dev/null && \
	ffmpeg -i VID_20220910_133415_5.mp4 -c:v libvpx-vp9 \
	-crf $NUM -b 0 -vf vidstabtransform -pass 2 -row-mt 1 -c:a \
	libopus VID_20220910_133415_5_ff.webm

And the result is probably better than Motion Plugin can do:
https://croatiafidelis.hr/foss/cap/cap-220925-ffmpeg/
(there's the original now too, the 221M	file)
https://croatiafidelis.hr/foss/cap/cap-220925-ffmpeg/VID_20220910_133415_5_ff.webm
The previous, the older is with NUM=40, the one posted above is
just .webm:

-rw-r--r-- 1 mr mr  72936332 2022-09-25 18:20
VID_20220910_133415_5_ff.webm

-rw-r--r-- 1 mr mr 127332623 2022-09-25 17:56
VID_20220910_133415_5_ff.webm_1664121928

( the SHA256 of it is:
c3cbbbb1ea23edd84edd8f704e610c096c2ea6d6c4b21b6b6b49ea84f07493e2
Cin/VID_20220910_133415_5_ff.webm )

And I should expect that the result can probably even be
improved.

Georgy Salnikov, thank you most sincerely again!

Regards!

-- 
Miroslav Rovis
Zagreb, Croatia
https://www.CroatiaFidelis.hr
my PGP-key:
https://www.croatiafidelis.hr/FCF13245ED247DCE443855B7EA9884884FBAF0AE.asc
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 659 bytes
Desc: not available
URL: <https://lists.cinelerra-gg.org/pipermail/cin/attachments/20220925/82d5da3e/attachment.asc>


More information about the Cin mailing list