Can't get MVAnalyseMulti or MVDegrainMulti Filters to work

[carlmart]

How do I get MVAnalyseMulti to work or MVDegrainMulti?

I have MVtools on my plugin, but AvsP refuses to accept any command involving these. Why is that?

[Gavino]

Which version of mvtools are you loading?

MVAnalyseMulti and MVDegrainMulti are from the 'Josey Wells' multi-threaded version of MVtools. The latest version is mvtools2.dll, where the functions are called MAnalyse and MDegrain2, etc.

[carlmart]

Which version of mvtools are you loading?

Version 1.11.4.5

MVAnalyseMulti and MVDegrainMulti are from the 'Josey Wells' multi-threaded version of MVtools. The latest version is mvtools2.dll, where the functions are called MAnalyse and MDegrain2, etc.

I also have MVtools2.dll there. I will try changing the words to see if it works now.

Continue: I am having problems making it work. Particularly because I don't know how to adapt the older parameters that were:

vectors= clip.MVAnalyseMulti(refframes=3, pel=2, blksize=16, overlap=8, idx=1)
denoised= clip.MVDegrainMulti(vectors, thSAD=800, SadMode=1, idx=1)

As I can't find any reference to MVAnalyseMulti and MVDegrainMulti I can't change what to what.

[Gavino]

I also have MVtools2.dll there. I will try changing the words to see if it works now.
OK. You might want to check the documentation that comes with mvtools2 (and look at the examples), as the parameters have changed from the older variants of mvtools.

[GMJCZP]

Please check this link:

http://forum.doom9.org/showthread.php?t=144271

And download Film_Restoring.zip. There's the version you want.

[cobo]

Continue: I am having problems making it work. Particularly because I don't know how to adapt the older parameters that were:

vectors= clip.MVAnalyseMulti(refframes=3, pel=2, blksize=16, overlap=8, idx=1)
denoised= clip.MVDegrainMulti(vectors, thSAD=800, SadMode=1, idx=1)

As I can't find any reference to MVAnalyseMulti and MVDegrainMulti I can't change what to what.

I think the MVTools2 version would be:
super= clip.MSuper(pel=2)
bv1 = MAnalyse(super, isb = true, delta=1,blksize=16,overlap=8)
fv1 = MAnalyse(super, isb = false, delta=1,blksize=16,overlap=8,)
bv2 = MAnalyse(super, isb = true, delta=2,blksize=16,overlap=8)
fv2 = MAnalyse(super, isb = false, delta=2,blksize=16,overlap=8)
bv3 = MAnalyse(super, isb = true, delta=3,blksize=16,overlap=8)
fv3 = MAnalyse(super, isb = false, delta=3,blksize=16,overlap=8)
denoised= clip.MDegrain3(super, bv1,fv1,bv2,fv2,bv3,fv3,thSAD=800)

Information about MVToolsMulti is interspersed in the MVTools thread.
The first mention of a multi version is here:
http://forum.doom9.org/showthread.php?p=1166868#post1166868
The last post by JoseyWells is here:
http://forum.doom9.org/showthread.php?p=1188389#post1188389

VideoFred's Film_Restoring.zip, which can be downloaded from the post GMJCZP provided the link for, contains the next to last version of MVToolsMulti - v.1.9.7.5 (9/11/2008 6:13PM)

I just noticed the OP is from 2009, but the info may be useful to others.

I've uploaded the last version of MVToolsMulti in case anyone wants to only download it.
MVTools_V1.9.7.7.rar 1.83MB (9/25/2008)
http://www.mediafire.com/?cm05eg4s0mctnt8

[cretindesalpes]

The latest modified MVTools2 versions included in the Dither package (http://forum.doom9.org/showthread.php?p=1386559#post1386559) have equivalent functions, although the syntax is a bit different:
sup = MSuper (pel=2)
vec = sup.MAnalyse (multi=true, delta=3, blksize=16, overlap=8)
MDegrainN (sup, vec, tr=3, thSAD=800)

[cobo]

Should that be tr=3 in your example?

Is there a limit to delta or tr in your modified MVTools2?

Could your version be modified to use separate SAD for each plane as in MVToolsMulti? It can prevent color from rapidly moving objects being smudged on backgrounds in some spots where the luma value happens to be the same.
http://forum.doom9.org/showthread.php?p=1176598#post1176598
I want to try something to possibly improve the MVDegrain results. Currently the total SAD is used to find the motion vectors and then the total SAD is again used to weight the Luma and Chroma planes in MVDegrain. I would like to change the MVAnalyseXXX function to store the total SAD and the SAD's for each Luma and Chroma plane in the vector data. Then use the SAD for each plane to degrain that plane instead of the total SAD.


http://forum.doom9.org/showthread.php?p=1176820#post1176820
OK, here is version 1.9.7.3
...
This also has the modification of of calculating the SadLuma, SadChromaU, SadChromaV and carrying that information through the vector data.

I also added a SadMode to MVDegrainXXX

SadMode=0 - Works as old method
SadMode=1 - Works with new method which may or may not be better

[cretindesalpes]

Indeed it's tr=3 (fixed now)

For the SAD-per-plane thing, It's interesting and I made a note about your request. I have lots of things to do or finish before so I cannot process it shortly, sorry for that.

[cobo]

I have no idea of the work involved, that you're willing to consider it eventually is more than I had reason to expect, thanks. A separate "limit_" and "thSAD_" each for U and V planes might be even better.

Is MDegrainN limited to 32 temporal frames maximum in each direction as in MVDegrainMulti?

[cretindesalpes]

The maximum tr value for MDegrainN is 128 (that's 256 frames counting both directions). Anyway I think memory will most likely be the main limit, as well as a possible loss of accuracy caused by the accumulation of dozens of values during the filtering stage.