I had started to write out a response along the lines of "the editor went overboard, changed the shape of the areolas, used too much cloning, and the differences are still inline with my original explanation", but it's always best to question your assumptions and do some due diligence, so I decided take the opportunity to procrastinate and have a closer look and check I wasn't overlooking anything. I went through the most complete sets; Lara, Makise, Makima, Shenhe and Citlai, and used some tools like ELA (error level analysis) on various photos to double check that there's no obvious signs of anything other than the typical editing hallmarks you'd expect in a glamour shoot.
So I didn't find nothing... and it forced me to reconsider a few things and change my response, but I didn't find any smoking guns either.
As I suggested before, there are tell-tale signs of editing. I think we all expect this from glamour photography, so no surprises there. There's one big glaring mistake in picture 31 of the Makise set, but I won't say what it is if anyone wants to take a crack at finding it themselves, it's hard to miss. However, there are signs of potential cloning and masking around specific areas consistently enough in pictures of earlier sets (Citlali, Shenhe, but mostly Makima) that make the possibility of prosthetics marginally more credible and got me to look deeper.
Now, before anyone gets too excited, there are not signs in every photo from these sets, but there are enough artifacts in the same spots (where shoulder straps would lie on the shoulders), in enough photos to warrant a closer look. I'm going to preface the rest of this by saying that I am a layman when it comes to forensic image analysis, I am by no means an authority and could very well be wrong, so my opinion with the appropriate grain of salt and degree of skepticism. I do have years of experience professionally editing, rendering, and creating digital imagery (both in a creative and technical capacity), so I do have insights the average person won't. This is by no means a thorough or conclusive analysis, it's just my observations and opinion after looking at most of the pictures available, and running a couple dozen through a variety of tools. With that out of the way, this is what I found:
- As mentioned before, I found consistent signs of potential cloning and masking along the edge of the shoulders, especially in the Makima set, with a profile that would match that of a strap (possibly to support a prosthetic), but only in some of the photos. Having said that, I couldn't find any sign of cloning or masking in any of the other areas you'd expect to find them were she wearing a prosthetic.
- What I did find can be explained as artifacting from a couple rounds of compression (this is why access to the original export is important). It's the fact that they were consistently in the same area over multiple photos that made me suspicious. Some do look a lot like bad cloning to me, but this could be confirmation bias since I'm looking for it. I'm not confident of what they are.
- Her photos appear to have monochromatic noise added (via a blending layer) in post, which is a method that could be employed in order to aid giving a prosthetic a more "skin" like texture and help hide cloning. HOWEVER, this is a very common method used in photo editing to give the picture a more grainy film like aesthetic, so it's probably nothing.
- Color-matching a prosthetic's hue over a gradient range to match skin is hard to do without leaving tell-tale signs and artifacts, especially at the terminator line. I couldn't find any obvious inconsistencies or signs of this.
- The only finds I had a hard time reconciling (in the Shenhe set) were terminator lines that look like they were poorly cloned, but this could simply be due to the high exposure of the sunlight coupled with image downsizing (they're not the original resolution).
- There are NO obvious signs of AI in the ELA passes, such as high levels of noise that's uniform across the entire image with higher densities on high contrast edges in all three color channels. The results of ELA are consistent with what you'd see from regular photography and editing. This doesn't rule AI out, but does make it extremely unlikely. AI is often easy to spot using ELA.
So I did find some anomalies that gives some weight to the possibility of a prosthetic. If they did use one, they've done a very good job of editing the typical tells out. However, there's also explanations that could adequately account for the presence of these anomalies, and the anomalies alone wouldn't be direct proof of a prosthetic, just signs of an edit that could account for one. This is assuming they are actually anomalies in the first place and it's not confirmation bias or an error on my part.
There are also many things that work against the notion of the prosthetic, many of which I didn't mention because I thought the last post was long enough as it is. The approximate size, proportions and shape of the breasts are well within the margin of any changes that might naturally occur over time (menstrual cycles, age, changes in weight, etc.). In my view, the most credible evidence for a prosthetic is on the basis of shape in side profile pictures of the older sets compared to newer ones; the nipples appear to be angled down more, the breasts have a bit more sag, and the overall shape of the breast is a bit more bulbous. However, all of these changes can reasonably be explained by changes in physique over time and/or by changes to editing. Also, the manner in which they change shape due to being squished or the pulled by gravity in different orientations (poses) also appears to be consistent with real breasts. Without access to the full catalogue of her work it's hard to say with any certainty, but judging by the photos available on Simpcity, you can see a gradual change to her breasts over the course of progressive sets (especially her nipples and areola).
So after looking into it a bit more, I think it's slightly more plausible that she could have worn a prosthetic in the earlier sets, but I still think it's more likely that the editor cocked up the execution and/or Blosmere is sensitive about the shape of her areolas and wanted them adjusted in earlier sets. Either way, I don't share your opinion that they're sufficiently different enough to have AI or a prosthetic be the only, or more likely explanation. I think my original critique and theory that it's just bad editing is still far more likely.
Again though, I'd ask why? Why would they bother in the first place, especially when the real thing is only marginally different and so much better? Without a plausible explanation for the motives behind the use of a prosthetic or AI being, only to be dropped for the real thing, it's hard to make sense of. I have yet to see any explanation offered for this and think it's required to make the notion plausible without hard evidence.
Look man, believe what you want to believe, it makes no difference to me. If you don't find my rebuttal convincing, I don't think there's going to be anything I could offer that would persuade you otherwise. Short of full analysis of the catalogue of original exports (which we don't have access to), it's hard to draw any conclusions with a high degree of confidence. That's why it's hard to take other people seriously when they confidently claim they know it's a prosthetics or AI, and all they have to offer in support of their claim is "look at this picture, it looks weird to me" and "bro, it's obvious, trust me". I think you guys are turning a molehill into a mountain and entertaining some pretty whacky ideas that can be simply explained by bad editing, but that's just my opinion, you can take it or leave it.
