If Gimp is no more what it used to be, maybe check out Krita. It is free, it is still pretty alive and it has a Python API:

https://scripting.krita.org/lessons/introduction 

Poser for some stupid reason cannot copy morphs from a prop into a figure. The solution is to make a one-bodypart figure with the prop geometry. Just use 'hip' as body part name. Then load the morphs from the original prop into the 'hip' of the new figure. From there you can copy into your target figure.

A tip about morph copy in shoes: do them left and right separately, and move left far out of the way when doing right and vice versa.  

Good to see you succeeded! At the time I tried to include a Prefitter for the male figures but I could not get it right. The heavy musculature in the base figures introduced, to my liking, too much noise in the morphs.

If this is her standard suit, you could use a rigged version of it as the base figure and add the head, boots/gloves, unless she does her superhero things barefoot and bare handed of course.

Saves a lot of pokethtu headaches and fixes.

As far as I understand @Iwanmtr has separate images for each block in the uv space, it is just that his geometry has uv coordinates in say the range (0,0) to (2,1).

AFAIK Poser deals with that seamlessly because it uses only the modulo of the coordinates.  

Yes, seems like the link system is broken.

Just for the fun of it I fed in this image 

which was an improvement made in Krea from a simple render of the Poser2 woman and LaFemme2. In the prompt I asked to let them dance to 'I can't get no satisfaction'. 

The resulting video is here.Here 

You can also use the information in the u- and v nodes in the Material room to block out areas of the texture.   

Alternatively,  if the area is not easy to describe mathematically you can use a greyscale and mathfunctions|step.

Problem: the preview render may not show this correctly and you need to (preview?)render to see the effects.

@unrealblue:  Fully agreed there! 

@Bejaymac: true, but again that all is just a matter administration. On loading the figure mesh Poser duplicates vertices when the facets they define are members of different groups that have the names of body parts. That's all there is to it. The problems are:

1 This subdivision is not transparent to the user. The .obj that is generated with the. cr2 is exposed and may overwrite the original. That is bad. Had it been named differently, say have extension .pfg (PoserFigureGeometry) and preferably be in a binary format nobody would have known or cared. 

2 Poser does not keep record of the relationship between the original figure mesh vertex number and body part mesh vertex number, or if it does, this is not exposed or utilised: this is again a matter of administration.  Note that the Python method 'unimeshinfo()' yields a weled mesh and a with full mapping of the relationships that is stable on reload.    

3 Same holds for the fbm's. Poser simply exposes more than it should do, and yes this clutters the parameter list to an irritating level. Exposure of parameters, again, is just an administrative thing.  

What makes her stand apart from the other figures you mention is that she is available for both Poser and DS and thus, in theory, it would be easier to make supporting items for her that serve both worlds. Rigging systems may be different but a geometry facet is a geometry facet, a morph delta is a morph delta and a texture map is a texture map. The other differences between Poser and DS systems are mere  administration and should be open to automatic conversion. 

Somehow however this is not capitalised on in the communications so I am not sure about the value in practice.  

The face is plain for easy character morphing. The body looks realistic and is easy to play with. She takes Victoria poses quite well.

Hmm. Looks like I have to build in a master dial that forces this one to 1 at dial value 0.

Correct about the dress one. But who is the beauty on the left?

I think AI could be of use for Poser in Poses. Poses are vectorized information, and there is plenty of it. It would only need a qualification that could link it to terms in the prompt. Problem, as ever with AI, is ownership of data.

As far as image enhancement is concerned I think it is better if Poser stays true to it's roots: provide template imagery/geometry of posed/animated figures and leave the final workout to other apps dedicated to that task. There is no way a small company like Bondware could ever stay par with development in the sector. 

In another thread on this subject I showed the below AI enhanced image:

Guess the Poser figures used.