AbsoluteReality 絕對現實

That feeling after you wake up from a dream

Add a :heart: to receive future updates. This took much time and effort, please be supportive :people_hugging:
Do you like what I do? Consider supporting me on Patreon :parking: or feel free to buy me a coffee :coffee:

Additional ecamples for V1.6: Image post by Lykon
Additional examples for V1: Image post by Lykon
Amazing gallery by qf22: Mixed with Vermeer painting
Quick face alteration examples using celebrity names and mixing them: Image post by Lykon

Available on sinkin.ai, Mage.space and many other services

Suggestions

  • Use between 4.5 and 10 CFG Scale and between 25 and 30 Steps with DPM++ SDE Karras. Worse samplers might need more steps.
  • To reproduce my results you MIGHT have to change these settings:
    • Set “Do not make DPM++ SDE deterministic across different batch sizes.” (mostly for v1 examples)
    • Set the ETA Noise Seed Delta (ENSD) to 31337
    • Set CLIP Skip to 2
    • DISABLE face restore. It’s terrible, never use it
  • Use simple prompts**.** Complex prompts might make less realistic pictures because of CLIP bleeding. More complex prompts does not mean better results. Keep it simple.
  • Use ADetailer to enhance faces. Basically every solo portrait I made uses it. You can get my settings by clicking on “copy generation data”. I suggest you use denoising under 0.3 to avoid getting always the same face.
  • Use BadDream and UnrealisticDream negative embeddings (BadDream, (UnrealisticDream:1.2)). Add weight to UnrealisticDream between 1.2 and 1.5. Do not use FastNegative or EasyNegative if you aim at realism. However, they’re good for artworks.
  • Use Highres.fix with the following settings: Denoising strength: 0.45, Hires steps: 20, Hires upscaler: 8x_NMKD-Superscale_150000_G and as much upscale as you can (my gpu only handles up to x1.8 at 512x768 base resolution, but you can go higher). If you don’t have 8x_NMKD-Superscale_150000_G you can probably use another GAN, but it should be easy to find on Google. You can also try Latent with a denoise higher than 0.6, but the result will be harder to control.
  • Try to condition faces by prompting for eye colors, hairstyles, hair color, ethnicity and so on. Even celebrity names do work. This model is pretty good at not making a single face if you play with the context.
  • If the pic is too clean, try to add some ISO noise. Even as a post processing with external tools it will trick the brain enough to make you think "damn, this is a real photo.
  • If you feel the grounded nature of this model is limiting your imagination, try generating on DS6 and then do img2img with this one to bump up the realism.

Pruned vs full comparison (not highres fixed)

Brief story of this model

While working on DreamShaper 6 I made many other models and a crap ton of tests. Some of them were strange mixes with photorealistic models I had previously made, plus new ones and some ISO noise LoRA I made. When I tested the various release candidates of DS6 for photography prompts, this came out on top. While mostly on the losing side with regards to range, flexibility and LoRA compatibility compared to what became DS6, I noticed this was pretty good at recreating photos with very simple and minimalistic prompts. So, why not, I gave it some love and kept working on it, just in time for its initial release.

Difference with DreamShaper

Long story short, DS aims at art, this aims at realism. They might overlap a bit, but they have different objectives and different things they’re good at. DreamShaper is total freedom and can basically do everything at a high level. This does about one thing and does it extremely well. This still uses DreamShaper as base, so it’s capable of doing art to a lesser degree.

絕對現實

那種從夢中醒來後的感覺

添加:heart:以接收未來的更新。這花了很多時間和精力,請支持 :people_hugging:
你喜歡我所做的嗎? 考慮在Patreon :parking:上支持我**,或者隨時請** 我喝杯咖啡:coffee:

V1.6 的其他示例: https : //civitai.com/posts/353121
V1 的其他示例 :https : //civitai.com/posts/259634
qf22 提供的畫廊: Mixed with Vermeer painting
使用名人姓名並混合它們的快速面部更改示例:Image post by Lykon

可在Sinkin.ai Mage.space和許多其他服務 上使用

建議

  • 使用 DPM++ SDE Karras 的 4.5 至 10 CFG 比例和 25 至 30 步。較差的採樣器可能需要更多的步驟。
  • 要重現我的結果,您可能必須更改這些設置:
    • 設置“不要使 DPM++ SDE 在不同批量大小上具有確定性”。(主要針對 v1 示例)
    • 將 ETA 噪聲種子增量 (ENSD) 設置為 31337
    • 將剪輯跳過設置為 2
    • 禁用面部恢復。太可怕了,千萬別用
  • 使用簡單的提示**。** 由於 CLIP 出血,複雜的提示可能會導致圖片不太真實。更複雜的提示並不意味著更好的結果。把事情簡單化。
  • 使用ADetailer增強面部效果。基本上我製作的每幅個人 肖像都使用它。您可以通過點擊“複製生成數據”來獲取我的設置。我建議您使用 0.3 以下的降噪,以避免始終得到同一張臉。
  • 使用BadDreamUnrealisticDream 負嵌入 ( BadDream, (UnrealisticDream:1.2) )。將權重添加到 UnrealisticDream 1.2 到 1.5 之間。如果您追求真實感,請勿使用 FastNegative 或 EasyNegative 。然而,它們非常適合藝術品。
  • 使用 具有以下設置的 Highres.fix:Denoising strength: 0.45, Hires steps: 20, Hires upscaler: 8x_NMKD-Superscale_150000_G 並儘可能高檔(我的 GPU 在 512x768 基本分辨率下最多只能處理 x1.8,但你可以更高)。如果你沒有,8x_NMKD-Superscale_150000_G 你可以使用另一個 GAN,但它應該很容易在 Google 上找到。您也可以嘗試使用高於 0.6 的降噪的 Latent,但結果會更難控制。
  • 嘗試通過提示眼睛顏色、髮型、頭髮顏色、種族等來調節臉部。即使是名人的名字也能發揮作用。如果你玩弄上下文的話,這個模型非常擅長不做一張臉。
  • 如果圖片太乾淨,請嘗試添加一些 ISO 噪點。即使使用外部工具進行後期處理,它也會欺騙大腦,讓您認為“該死,這是一張真實的照片。
  • 如果您覺得該模型的基礎性質限制了您的想像力,請嘗試在 DS6 上生成,然後使用此模型執行 img2img 以提高真實感。

修剪與完整比較(不固定高分辨率)

該模型的簡要故事

在使用DreamShaper 6時,我製作了許多其他模型和大量測試。其中一些是我之前製作的逼真模型的奇怪混合,加上我製作的新模型和一些 ISO 噪聲 LoRA。當我測試 DS6 的各個候選版本的攝影提示時,它名列前茅。雖然與 DS6 相比,它在範圍、靈活性和 LoRA 兼容性方面處於劣勢,但我注意到它非常擅長通過非常簡單和簡約的提示來重新創建照片。所以,為什麼不呢,我給了它一些愛並繼續努力,正好趕上它的首次發布。

與 DreamShaper 的區別

長話短說,DS的目的是藝術,這目的是現實主義。他們可能會有一些重疊,但他們有不同的目標和擅長的事情。DreamShaper 是完全自由的,基本上可以以高水平完成所有事情。這只做一件事,而且做得非常好。它仍然使用 DreamShaper 作為基礎,因此它能夠在較小程度上進行藝術處理。