Stable Diffusion是一种利用深度学习技术进行图像创造的神奇艺术，而在这些图像诞生的过程中，风格提示词（prompt）就像是一位操控舞台的魔法师，发挥着举足轻重的作用。本系列特别为您收集整理了100多种Stable Diffusion风格提示词，我们将通过SDXL一一测试，并将每一种风格的效果图展现出来，让您一目了然。敬请期待这...
Prompt: Aesthetic pastel magical realism, a manwith a retro TV for a head, standing in the centerof the desert, vintage photo. 图源：X 调用API，网友用SD和ChatGPT做了一个小游戏。图源：X 在他看来，SD3能比DALL·E有更好的风格一致性，对于生成一个系统工程中的美术素材表现得更好。网友也分享...
Prompt: Portrait photograph of ananthropomorphic tortoise seated on a New York City subway train. 对于完全虚构的图片也能做到美学和细节真实度的统一。 Prompt: Aesthetic pastel magical realism, a manwith a retro TV for a head, standing in the centerof the desert, vintage photo. 调用API,网友用SD3...
除了通过本地下载模型调试代码,stable diffusion还有很多开源的服务,便于做快速测试。 以https://huggingface.co/spaces/stabilityai/stable-diffusion 为例,同样可以输入一句prompt生成图像,不过由于资源紧张需要排队使用,排队时间也可能比较长。 这里同样测试了几个prompt,效果如下: a photo of a smiling Chinese women ...
在 Stable Diffusion 中，它被用作概率编码器（Encoder）和解码器（Decoder）。VAE 通过将输入数据映射到潜在空间中进行编码，然后将编码的向量与潜在变量的高斯分布进行重参数化，以便可以直接从潜在空间中进行采样。2. Unet 是一种基于卷积神经网络的图像分割模型，采用了一种特殊的“U”形结构，使得输入的分辨率逐步...
1), dieselpunk style, (Kou Yokoyama style:1. 2), Concept art top Artstation, masterpiece, combat action in a ruined city in the background, retro-futuristic style, (rounded shapes:1. 2), mechanisms, maximum detail, vintage, smoke, dramatic background, realism, digital painting, dark ...
1. Suggest a prompt for an mysterious house Request ### Human:suggest a prompt for an mysterious house ### Assistant: Mysterious house, haunted mansion, spooky, intricate, sharp details, octane render, unreal engine, photograph, realistic skin texture, photorealistic, hyper realism, highly detaile...
Prompt: Portrait photograph of ananthropomorphic tortoise seated on a New York City subway train. 图源:X 对于完全虚构的图片也能做到美学和细节真实度的统一。 Prompt: Aesthetic pastel magical realism, a manwith a retro TV for a head, standing in the centerof the desert, vintage photo. 图源:X ...
These styles are the presets in DreamStudio, the official Stable Diffusion generator. 3D Model Prompt professional 3d model {prompt} . octane render, highly detailed, volumetric, dramatic lighting Negative prompt ugly, deformed, noisy, low poly, blurry, painting ...
Stable Diffusion (DreamShaper) Easy to generate good images Midjourney is well-known for being surprisingly easy to generateartistic images with a lot of fine details. You don’t need to work very hard to generate good images. If fact, very often, it will ignore part of your prompt and ...