As you all know, as a skinsuit content creator, I've always been very particular about the details and realism of the skinsuit effects. That's why I've always wanted to create extremely realistic and delicate skinsuit textures — including wrinkles when the suit is stretched, visible thickness at the edges of the skinsuit, seeing the other person underneath through the stretched and torn-open parts, the tearing sensation and fleshy feel when ripping the skinsuit apart, sweat and slime beneath the skinsuit, and many other such details.
However, I'm limited by the capabilities of 3D rendering software, so I can't achieve these effects when rendering my comics. The workload is simply too heavy for me to handle alone.
But at the same time, the emergence of AI has given me a glimpse of feasibility. I’ve discovered that as long as the prompt is well-crafted, AI can generate many of the effects I want to see — for example, the video clips and GIF animations I posted here. These can, to a certain extent, satisfy my pursuit of realistic skinsuit effects. If I can successfully incorporate such effects into my comics in the future, it will surely elevate the visual quality and appeal of my work.
However, AI still has many limitations at present. My local models can handle NSFW content, but due to their parameter constraints, the video quality is far inferior to what online platforms can generate. On the other hand, online platforms have extremely strict content censorship, making it very difficult to generate more explicit scenes. So right now, I’m stuck in a situation where I can’t have both. I need some better ideas on how to combine the strengths of both approaches.
At the same time, because my time and energy are limited, I don’t have the time to figure out how to optimize video prompts. Also, there are already many people using AI to create skinsuit videos and stories, so I don’t plan to shift my main focus to AI generation. I mainly hope to use AI to improve the visual quality of my comics and to provide accompanying videos — similar to what Carnal Weaver 2, 3, and The Skin Auction did.
That said, if I get good AI results, I still hope to share them with everyone. I also welcome any suggestions from you all. If any of you are also using AI to create content, feel free to discuss model and prompt-related issues with me!
I’ve also sent the video links in private messages and the chat group. Please download them yourself.
你们知道，作为一个皮物内容创作者，我一直很追求皮物效果的细节和真实度，所以我一直想做出非常真实细腻的皮物质感，包括拉扯时候的褶皱，皮物边缘能够体现皮物厚度，透过拉扯开的皮物能够看到皮下的另一个人，还有撕开皮物时候的撕裂感，血肉感，皮物下面的汗水，粘液等等这些细节。
但是受限于3D渲染软件的功能，我没法在渲染漫画的时候做出这样的效果，因为工作量实在是太大了，我一个人没法做到。
但与此同时，AI的出现让我看到了一些可行性，我发现只要提示词合适，就能让AI生成许多想要看到的效果，比如我列出的这些视频瞬间以及GIF动图，可以一定程度上的满足对皮物效果的追求，如果后续我能将这样的效果用在漫画中，想必可以提升我漫画的观感。
不过目前AI的限制还是很多，我本地的模型可以处理NSFW画面，但是因为本地模型的参数限制，视频的效果远远比不上在线平台的生成，而在线平台对内容审查极为严格，很难生成比较露骨的画面，所以目前还处于不可兼得的阶段。我需要一些更好的思路来结合起来。
同时因为我的时间精力有限，没有时间去琢磨怎么优化视频的提示词，而且目前也有很多人在通过AI制作皮物类的视频和故事了，所以我并不打算把重心转移到AI生成上，主要还是希望通过AI提高我漫画的质感，以及提供配套的视频，就像Carnal Weaver 2，3以及Skin Auction那样。
不过有好看的AI结果我还是希望与大家分享，也希望大家能够给我提一些建议，如果也有在用AI制作内容的，也欢迎与我讨论模型和提示词方面的问题！
视频链接也发在了私信和chat group中，请自行下载。