#ADetailer

[munchsstore]

A Symphony of Beauty Celebrating Heritage and Tradition with AI 👰‍♀️👗💍✨ @munshsstore Description: Step into a world where couture becomes an exquisite celebration of the world's cultural tapestry. In this captivating video, witness how garments transcend their utilitarian purpose and transform into breathtaking art pieces that blend traditions and heritage. Immerse yourself in a high-quality image rendered with cinematic brilliance, dynamic lighting, and meticulous attention to detail.

[aixmedia]

Hellaine Illustration

Merge/Partial train using Hellaine Pony XL Lora Base Model used: PonyXL/Osenayan Animation PonyXL/ AutismConfetti Mix Where to get: https://civitai.com/models/362212/hellaine-illustration-xl

Join us on our journey Website: https://www.end-media.org Discord: :https://discord.gg/5t2kYxt7An Backups: https://huggingface.co/EarthnDusk Send a Pizza: https://ko-fi.com/duskfallcrew/

About & Sponsored

"WE"? - We have Dissociative identity disorder, ADHD, Autism and CPTSD - "WE" as in we're a system of over 200 alters, and we're not ashamed about it. We believe that AI can break down barriers in some aspects of mental health, but we also believe that AI can hinder aspects of it.

JOIN OUR DA GROUP: https://www.deviantart.com/diffusionai

JOIN OUR SUBREDDIT: https://www.reddit.com/r/earthndusk/

HOW TO USE:

EXTENSIONS:

We use ADETAILER: https://github.com/Bing-su/adetailer.git (BUT NOT THAT OFTEN ON SDXL, VERY SPARINGLY)

DUSK DO YOU USE CONTROLNET- YES SOMETIMES BUT I DONT HAVE A TUTORIAL YET - AND IM TOO LAZY.

Batchlinks if you're using A1111 locally or on a server: https://github.com/etherealxx/batchlinks-webui @nocrypt Addon: https://github.com/NoCrypt/batchlinks-addon

[whisper-and-tangle]

ADETAILED ADN POETIC THREAT 2 TGHE PERSON WHO INTENTIONALLY VANDALIZED MY 1994 DODGE INTREPID BEHIND KATE'S APARTMENT MY BELOVEDD<3333

[astarionsas]

What program/AI do you use? No hate, I just wanna know… asking for a friend. 👀

Stable Diffusion version: v1.6.0 with ADetailer for faces. Loras I found here https://civitai.com But also I use photoshop for every pic My 1060 is not enough for a good result

[red-pixel]

After creating my first Lora (based on my LARP Photos) for SD1.5 and SDXL some more polished Images. Used ProtovisionXL for the composition. Later Img2Img with 1.5 CardosAnime with adetailer and 0.42 denoise.

[wdmcblog]

Black succubus warrior

Prompt: (best quality, masterpiece :1.3), (beautiful succubus warrior), (black leather armor, black leather shoulder pads, leather baldric, black leather gloves, black leather skirt, :1.2), (sexy, beautiful, perfect breasts, medium breasts :1.2), (pale white skin:1.3), Portrait of 20yo succubus, (sexy pose :1.2), detailed face, Necklace, beautiful eyes, succubus walking on the desert land, scorched wasteland, with the mesmerizing expanse of outer space as her backdrop, while the Moon, fantasy theme, beautiful and aesthetic, official art, (full body), long shot, perspective view, small_horns, red hair, wavy hair, wind, (detailed white eyes), detailed and realistic, ArtStation, blending the artistic styles of Luis Royo, Julie Bell, and Brom, in frame, natural skin, textured skin, (green_eyes :0.6) Negative prompt: ( raincoat, cape, wings, sword :1.2), (worst quality:1.6), (low quality:1.6), (normal quality:1.6), lowres, bad anatomy, bad hands, vaginas in breasts, ((monochrome)), ((grayscale)), collapsed eyeshadow, multiple eyebrow, (cropped), oversaturated, extra limb, missing limbs, deformed hands, long neck, long body, imperfect, (bad hands), signature, watermark, username, artist name, conjoined fingers, deformed fingers, ugly eyes, imperfect eyes, skewed eyes, unnatural face, unnatural body, error, painting by bad-artist Steps: 60, Sampler: DPM++ 3M SDE Karras, CFG scale: 7, Seed: 80470869, Size: 768x1024, Model hash: 5ebd6720d4, Model: zavyyumexl_v10, VAE hash: 551eac7037, VAE: sdxl_vae.safetensors, Denoising strength: 0.35, ADetailer model: face_yolov8n.pt, ADetailer confidence: 0.3, ADetailer dilate/erode: 4, ADetailer mask blur: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding: 32, ADetailer version: 23.8.0, Hires upscale: 2, Hires steps: 60, Hires upscaler: 4x_NMKD-Siax_200k, Lora hashes: "add-detail-xl: 9c783c8ce46c", Version: v1.6.0

[seriously-mike]

So I decided to recreate a concept from last year using Flux. On the left, horribly overcooked output of a third-party Stable Diffusion 1.5 dataset (Protogen Infinity) with an embedding of Keira Knightley's face, edited in Photoshop to mask out the arm and edit the nametag, then shoved back into Stable Diffusion to add the cybernetic arm. 120 steps in total, split into two runs: 80 to draw the base and 40 to inpaint the arm.

On the right, the result of an overengineered workflow involving Flux Dev for drawing the base and, because Flux doesn't know how celebrities look, ADetailer plugin to automatically mask out the face every time and a third-party SD 1.5 dataset (Dreamshaper 7) to redraw it using the same embedding I used last year (which is a last year trick, but with an additional plugin used for precision). 47 steps in total, one run and no Photoshop involved.

Also, a bonus round, because sometimes Flux still fucks up or confuses left and right. Hyper's "Spoiler" intensifies.

[welcometoparallelworlds]

Belldandy in Alphonse Mucha style🤭 I did not include ADetailer to fix hands in this batch so hands were not quite right in many of this series...

[ifotoeditor]

The Best Alternative Tool for ADetailer in 2024: One-Click Face Swapping!

The progress of image editing technology have opened a new creative world for us. With the continuous development of software tools, we can now process and enhance photos and images in unprecedented ways. Modern image editing software provides a wide range of functions from basic color adjustment to complex image synthesis, which can meet the needs of everyone from amateurs to professional designers. Face swapping technology, in particular, has made a splash not only in entertainment and social media but also in film production, game development, and even news reporting. This technology enables the creators to make convincing visual effects and completely fictional scenes by replacing faces in images. ADetailer is such tool, which simplifies the process of changing faces by using artificial intelligence algorithms, so that users can realize high-quality image editing without complicated technical knowledge.

As these tools becoming more and more intelligent and user-friendly, we can foresee that in the near future, image editing will become more intuitive and creative, providing users with more possibilities and freedom.

The best way to use LoRA: ADetailer+LoRA Face Swapping

Install the ADetailer plug-in: Search for ADetailer in the plug-in tab of Stable Diffusion. If you cannot find it, you can use the "Install from URL" option to install it through the project URL on GitHub. The process is simple and straightforward, just copy the URL from the GitHub page, then paste it in the WebUI and hit install.

Download and install the model: Depending on your needs, you may need to download a specific model from the Internet and move it to the folder specified by Stable Diffusion. This step ensures that the ADetailer plug-in can smoothly call the required model to perform the face swap operation. ModelTargetface_yolov8n.ptFace (Illustration/Real)face_yolov8s.ptFace (Illustration/Real)hand_yolov8n.ptHand (Illustration/Real)person_yolov8n-seg.ptPerson (Illustration/Real)person_yolov8n-seg.ptWhole Body (Illustration/Real)person_yolov8s-seg.ptWhole Body (Illustration/Real)person_yolov8s-seg.ptWhole Body (Illustration/Real)mediapipe_face_fullFace (Real)mediapipe_face_shortFace (Real)mediapipe_face_meshFace (Real) 3. Enable ADetailer: Find and enable the ADetailer option in the Stable Diffusion interface. This usually involves checking a box to activate the plugin's functionality.

4. Choose the right model: ADetailer offers a variety of models, including models dedicated to faces, hands, or entire figures. Depending on the type of image you want to process (such as a portrait or a full-body photo), choosing the right model will help improve the quality and accuracy of the face swap effect. 5. Adjust parameters: ADetailer provides a variety of parameters for you to adjust to more precisely control the results of face swapping. You can adjust these parameters as needed, such as changing the size or position of facial features. 6. Generate pictures: After setting all parameters, you can generate new pictures and observe the face swap effect. If the effect is not satisfactory, you may need to go back to the parameter settings and adjust it until you achieve satisfactory results.

The Best ADetailer Alternative to Swapping Face in One Click - iFoto Face Swap

Do you feel that ADetailer is too complex and professional to bring you the function of simplifying drawing? This is not an efficient and intuitive image editing tool. But that's okay! iFoto, provides users with powerful features that enable them to create and share easily. iFoto Face Swap function is a typical example. It allows users to quickly replace faces through a simplified operation process, whether for entertainment or to create more professional visual effects. Next, I will introduce in detail how to use iFoto Face Swap to help you better utilize this technology to enrich your picture creation. 1. Open iFoto Face Swap: After open iFoto, you will see an intuitive interface with various tools and options. Here you can browse and select images from your gallery, or simply snap a new photo to edit.

2. Select a picture: In the dashboard, you can select any picture stored on your device. If your goal is face replacement, it's best to choose a photo with clear foreground and background and even lighting to make editing easier. 3. Find the face swap feature: In the editing tools, find the face swap feature, which may be labeled "Face Swap," "Face Replacement," or "Face Fusion." This feature is usually found under the "Editing" or "Special Effects" category. 4. Upload a face picture: You need to upload a picture containing the face you want to replace. It is best to choose pictures with natural expressions and high resolution to ensure that the face swap effect is natural and seamless.

5. Adjustment and Alignment: Using the adjustment tools provided by iFoto, you can move, scale and rotate the new face to ensure that it matches the position and size of the original face. This step is critical to achieving a sense of realism. 6. Apply changes: After completing all adjustments, you can preview the face swap effect. If everything looks good, you can apply the changes. If you're not satisfied, you can undo your changes or readjust. 7. Save or Share: Once you are satisfied with the final result, you can save the edited image to your device or share it to social media platforms directly from iFoto One of the key strengths of iFoto Face Swap is its versatility, allowing users to seamlessly swap faces in their images with just a few simple clicks. Whether it's creating humorous memes, personalized greeting cards, or imaginative digital artwork, iFoto provides endless opportunities for users to unleash their imagination and bring their ideas to life. Additionally, the app's extensive library of pre-designed templates and effects further enhances the editing experience, providing users with a wealth of creative options to explore.

FAQs

1. Which devices support iFoto Face Swap? iFoto Face Swap is well-supported across a variety of devices, including both Android and iOS platforms. This means that whether you're using a Samsung, Huawei, Xiaomi, or an Apple phone, you can find and use iFoto from the respective app stores. This broad compatibility ensures that most smartphone users can enjoy the face-swapping fun of iFoto. 2. Is an internet connection required to use iFoto Face Swap? The basic face-swapping feature of iFoto Face Swap can be used offline without an internet connection. This is convenient for users who wish to edit photos without network access. However, some advanced features, such as cloud storage or specific image processing services, may require online access to use. 3. How to choose the right photo for face swapping in iFoto? Choosing the right photo is crucial when face swapping in iFoto. The ideal photo should be a high-resolution, well-lit frontal portrait. Avoid using blurry or poorly lit photos, as these factors can affect the quality of the face swap. Additionally, ensure that the face to be replaced is not obscured, allowing the AI to more accurately identify and process facial features. 4. Is the face-swapping process in iFoto instantaneous? The face-swapping process in iFoto is usually very fast, with users able to see results almost immediately in most cases. This is thanks to the advanced AI technology used by iFoto, which can quickly analyze facial features and perform replacements. Even when dealing with larger or more complex images, users only need to wait a few seconds. 5. Is it safe to use iFoto Face Swap for face swapping? iFoto takes user privacy and security very seriously. All face-swapping operations are performed on the user's device and do not involve uploading pictures to servers. Additionally, iFoto uses encryption technology to protect user data, ensuring that users' privacy is not compromised.

In Summary

iFoto Face Swap stands out as a comprehensive and user-friendly tool for photo editing enthusiasts of all skill levels. Its intuitive interface and robust functionality make it easy for users to explore and experiment with various face-swapping techniques, from subtle adjustments to dramatic transformations. With a diverse range of features and customization options, iFoto Face Swap empowers users to unleash their creativity and express themselves in unique and captivating ways. Furthermore, iFoto Face Swap's commitment to user satisfaction is evident in its regular updates and enhancements, ensuring that users always have access to the latest features and improvements. Whether you're a seasoned photographer looking to add a playful touch to your images or a casual user seeking to spice up your social media posts, iFoto Face Swap offers the tools and resources you need to elevate your photo editing game. With its user-friendly interface, powerful features, and endless creative possibilities, iFoto Face Swap is truly a must-have tool for anyone looking to take their photo editing skills to the next level. Read the full article

[rilmiah]

続・画像生成AI

本当にありがたい存在です。細かいところは無視してくるし、まったく作れないものも色々ありますが、それでもこの1か月で既に100くらいのキャラに画像を付けることができました。

Firefly、Pix.Ai、Tensor.Artそれぞれの、個人的に感じた特徴と比較。

Fireflyは人間が弱い。極々稀に、そのまま使えるものを出してくれますが、後は手がなかったりグチャグチャだったりというものを平気で掲載してきます。強調もネガティブプロンプトもないので「最低限従って欲しいもの」「出て欲しくないもの」を教えることはできず。ただ、一番モノを知っていて、一番「それっぽいもの」を出してくれる気はします。そして、回数制限なく何度でも試せるのもありがたい。大抵の場合、まずEmEditorでプロンプトを作ってFireflyで何度か試してから、PixまたはTensorにそのプロンプトを持って行くという使い方をしています。英単語のミスなんかの無駄打ちを減らせる。

Pixはモデルの選択が悪くなければ「思い切りハズレ」はそう多くありません。人間の顔を一番きれいに出してくれるので、リアル風味の美形・美少女を描かせるなら、この3つの中では一番だと思います。ただしみんな同じ顔で、年齢も比較的低めです。女性への偏りが大きくて男性を指示したのに女性が出てくることも多い。そして（1,000ptの高優先度を付けないと）待ち時間が長い。30分以上待たされることもある。毎日固定で10,000pt配布されて、4枚出力なら800pt消費なので、無駄打ちが少なければ0にはならない（というか待ち時間が長いので、「これはPixの美少女でないと」というキャラ以外はどうしてもTensorメインになる）。モデルを変えるとネガティブプロンプトがリセットされるのが面倒くさい。

Tensorはモデル数が多すぎかつ選択UIが悪いので毎回「どれだっけ…」と何回か無駄打ちしてしまいますが、Pixよりは絵柄の幅が出せます。しかしやはりモノを知らない。そして、顔が潰れてしまう確率が結構高い。ADetailerという顔修復専用のフィルタが最初から付いているのが、提供側としても問題点を把握してるからでしょう。とはいえ、ADetailerを1回、多くて2回通せばほぼ見られる顔になるので、そう大きな問題ではありません。どちらかというと、ネガティブで指定しても手が3本になったり指が4本/6本になったりが多い方が、修復できない分面倒くさい。毎日使えるのは100ptカッキリ（翌日消費分が補充される）しかないので、無駄打ちが多いとあっという間に使い切ります。

このうち、エルフの耳が一番期待どおりなのはFireFlyです。継いでモデルによってはTensorも割と。

[promptreferencesposts]

https://civitai.com/images/841111?modelVersionId=75209&prioritizedUserIds=747825&period=AllTime&sort=Most+Reactions&limit=20 Best quality, masterpiece, ultra high res, (photorealistic:1.4), raw photo, 1girl, white dress, off shoulder, blossom flower field, glowing skin, light smile Negative prompt: ng_deepnegative_v1_75t, (badhandv4:1.2), (worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, bad hands, ((monochrome)), ((grayscale)) watermark Size: 512x768, Seed: 3138559612, Model: majicMIX_sombrev2-fp16-no-ema, Steps: 30, Sampler: Euler a, CFG scale: 10, Clip skip: 2, Model hash: 5c9a81db7a, Hires steps: 25, Mimic scale: 7, Hires upscale: 2.5, ADetailer conf: 30, Hires upscaler: 4x-UltraSharp, ADetailer model: face_yolov8n.pt, ADetailer version: 23.5.6.post0, ADetailer x offset: 0, ADetailer y offset: 0, Denoising strength: 0.2, ADetailer CFG scale: 7.0, ADetailer mask blur: 4, Threshold percentile: 100, ADetailer dilate/erode: 0, ADetailer inpaint full: True, ADetailer inpaint width: 512, ADetailer inpaint height: 512, ADetailer inpaint padding: 0, ADetailer denoising strength: 0.1, Dynamic thresholding enabled: True, ADetailer use inpaint width/height: True

[squarwell-breakingnews]

Greater Noida West Metro gets approval from Central Govt, construction to begin SOON

Greater Noida West Metro gets approval from Central Govt, construction to begin SOON

The Noida-Noida Extension Metro Project that is an extension of the Aqua Line Metro in Uttar Pradesh’s Noida city and passes through the Greater Noida West area has finally received approval from the central government after much delay. The project was sent to the Public Investment Board (PIB) by the Ministry of Urban Development after adetailed project report (DPR) by the Uttar Pradesh…

[wdmcblog]

Red succubus warrior

Prompt: (best quality, masterpiece :1.3), (beautiful succubus warrior), (red leather armor, red leather shoulder pads, leather baldric, red leather gloves, red leather skirt, :1.2), (sexy, beautiful, perfect breasts, medium breasts :1.2), (pale white skin:1.3), Portrait of 20yo succubus, (sexy pose :1.2), detailed face,  Necklace, beautiful eyes, succubus walking on the desert land, scorched wasteland, with the mesmerizing expanse of outer space as her backdrop, while the Moon, fantasy theme,  beautiful and aesthetic, official art, (full body), long shot, perspective view, black hair, wavy hair, wind, (detailed white eyes), detailed and realistic, ArtStation, blending the artistic styles of Luis Royo, Julie Bell, and Brom, in frame, natural skin, textured skin,  <lora:add-detail-xl:0.8> Negative prompt: ( raincoat, cape, wings, sword :1.2), (worst quality:1.6), (low quality:1.6), (normal quality:1.6), lowres, bad anatomy, bad hands, vaginas in breasts, ((monochrome)), ((grayscale)), collapsed eyeshadow, multiple eyebrow, (cropped), oversaturated, extra limb, missing limbs, deformed hands, long neck, long body, imperfect, (bad hands), signature, watermark, username, artist name, conjoined fingers, deformed fingers, ugly eyes, imperfect eyes, skewed eyes, unnatural face, unnatural body, error, painting by bad-artist Steps: 60, Sampler: DPM++ 3M SDE Karras, CFG scale: 7, Seed: 4125954075, Size: 768x1024, Model hash: 5ebd6720d4, Model: zavyyumexl_v10, VAE hash: 551eac7037, VAE: sdxl_vae.safetensors, Denoising strength: 0.35, ADetailer model: face_yolov8n.pt, ADetailer confidence: 0.3, ADetailer dilate/erode: 4, ADetailer mask blur: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding: 32, ADetailer version: 23.8.0, Hires upscale: 2, Hires steps: 60, Hires upscaler: 4x_NMKD-Siax_200k, Lora hashes: "add-detail-xl: 9c783c8ce46c", Version: v1.6.0

[stylearche]

[seriously-mike]

The Perks of Being Open-Source

As promised, here's something about the plug-ins for Stable Diffusion, the only AI image generator that's open-source, and makes for a great testbed for related research because of that.

The outcome of this research are various plug-ins and modifications that allow Stable Diffusion to do a lot of interesting things. And the authors? Hooboy, you would be surprised. It turns out that very serious companies, ones you would suspect of cooking their own tech of this kind on the side even if they haven't announced it publicly, build interesting things and share it for free.

Let's start with Semantic Segmentation. If an AI image generator makes images based on text input, Semantic Segmentation scans images, identifies elements and assigns text descriptions to them. A lot of serious companies have released open-source code of their implementation: you have Nvidia, Google, Meta and even Alibaba building that stuff. It might sound kinda underwhelming if fairly useful for helping visually impaired people (for example, Facebook uses it to generate alt texts for images posts automatically), but here's the kicker: Semantic Segmentation may be used in Stable Diffusion to automatically generate masks based on text description. Want to find a hand and redraw it to be more anatomically correct? Easy. How about changing the hair color without monkeying with it in Photoshop? Also easy. So easy that some basement-dwelling chud can script it to find the clothes on a woman and draw a nude body in their place (and did, and got himself in trouble when another teenage chud uploaded photos of girls from his class to the app).

Semantic Segmentation is the core of the popular Stable Diffusion extension called aDetailer: as I mentioned before, Semantic Segmentation can recognize what a hand is, even if it's distorted, and point the generator to inpaint a better version in its place. Same goes for the faces. And that's the two things aDetailer is built to fix.

Another thing are ControlNets: plugins that allow you to nudge the generation process a particular way, be it recreating the pose of a character down to hands and fingers (or just the face orientation and expression), following the outline of a sketch and filling in the details, even maintain perspective using depth maps. And then, based on that tech, you have PhotoMaker, created by Tencent's Applied Research Center, and its improved version, IP Adapter. The capabilities are impressive, particularly if you remember that a slightly outdated gaming PC can run Stable Diffusion at a decent pace with no need for an internet access, even with the plugins.

Also, with OpenAI's video generator Sora looming on the horizon, you should know that the first AI-generated (or at least redrawn) videos were created in Stable Diffusion as well. I don't intend to go down this rabbit hole for practical reasons (I have no need for using it for that particular purpose and my video card is a bit outdated), but it was on the sweaty basement-dwelling nerds to figure out how to fine-tune the whole thing to be consistent across a whole fuckton of frames, and they did it, the crazy sonsabitches.

So laugh all you want at the ornery, wobbly Stable Diffusion producing rounded, fractal blorps and fucky hands. Even basic capabilities like inpainting and outpainting still make Midjourney jealous, and if you look at the plugins, you can imagine a good few use cases you could never wring out of the competing algorithms - and run them on your own PC for free instead of relying on centralized black boxes with a monthly fee.