AI drawing is quite popular recently, these are actually produced by Stable Diffusion, an AI drawing tool. However, for lazy people who don’t have much time to research and want to play with iPhone to pass the time easily, this time I would like to share with you this easy-to-install Google Colab cloud service through a browser, which can be installed on iPhone, Android phones, tablets, computers, etc. The way to generate pictures with one click.
If you want to play AI drawing easily, you can generate it with one click with your iPhone! (teaching)
This time, the author shares the lazy skills of using the iPhone to generate AI drawings with one click (for computer version operation teaching, please refer to Ada’s article), first use the Chrome browser to enter the URL: “https://colab.research.google.com/”, Then you need to log in to your Google account to proceed to the next step.
Then click “Connect” in the upper right corner of the screen, a message of connecting will appear at this time, and finally the status of “Connected” will appear and RAM and disk information will be displayed.
After clicking the location of RAM and disk, turn the screen to “Change Execution Phase Type” at the bottom, change the original hardware accelerator to “GPU”, and click the “Save” button.
Then go back to the original Colab page, find the “+” button in the upper left corner of the screen, find the dialog box and paste the installation instructions (refer to the instructions below).
Paste the following installation instructions and click the button to start the installation steps:
!pip install –upgrade fastapi==0.90.0
!git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui
!git clone https://github.com/yfszzx/stable-diffusion-webui-images-browser /content/stable-diffusion-webui/extensions/stable-diffusion-webui-images-browser
!curl -Lo chilloutmixni.safetensors https://huggingface.co/nolanaatama/chomni/resolve/main/chomni.safetensors
!curl -Lo ulzzang-6500.pt https://huggingface.co/nolanaatama/chomni/resolve/main/ulzzang-6500.pt
!curl -Lo koreanDollLikeness_v10.safetensors https://huggingface.co/duthanhduoc/chilloutmix-set/resolve/main/koreanDollLikeness_v10.safetensors
!mkdir /content/stable-diffusion-webui/models/Lora
!mv “/content/koreanDollLikeness_v10.safetensors” “/content/stable-diffusion-webui/models/Lora”
!mv “/content/chilloutmixni.safetensors” “/content/stable-diffusion-webui/models/Stable-diffusion”
!mv “/content/ulzzang-6500.pt” “/content/stable-diffusion-webui/embeddings”
%cd /content/stable-diffusion-webui
!git checkout 91c8d0d
!COMMANDLINE_ARGS=”–share –disable-safe-unpickle –no-half-vae –xformers –reinstall-xformers –enable-insecure-extension-access” REQS_FILE=”requirements.txt” python launch.py
Then wait for a while to install, until the URL appears, it means that the installation is complete. At this point, click the URL to open the Stable Diffusion web page. At this time, two explanatory boxes will be seen on the screen. The “Prompt” on the top indicates what the calculated image looks like, and the “Negative prompt” indicates what the calculated image does not want to look like. The suggested parameters will also be shared below for beginners who want to try for the first time.
Prompt parameter reference (please paste the following parameters in the Prompt field):
(8k, RAW photo, best quality, masterpiece:1.2), (realistic, photo-realistic:1.37), ultra-detailed,full body,1 girl, solo,beautiful detailed sky,detailed tokyo street,night,beautiful detailed eyes,beautiful detailed lips,professional lighting, photon mapping, radiosity, physically-based rendering,extremely detailed eyes and face, beautiful detailed eyes,light on face,cinematic lighting, short jacket,hoodie,school uniform,1girl,full body,full-body shot,see-through,looking at viewer,outdoors,
Negative Prompt parameter reference (please paste the following parameters in the Negative Prompt field):
EasyNegative, paintings, sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, age spot, glans,extra fingers,fewer fingers,strange fingers,bad hand,signature, watermark, username, blurry, bad feet,bad leg, duplicate, extra limb, ugly, disgusting, poorly drawn hands, missing limb, floating limbs, disconnected limbs, malformed hands, blurry,mutated hands and fingers,, EasyNegative, paintings, sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, age spot, glans,extra fingers,fewer fingers,strange fingers,bad hand,signature, watermark, username, blurry, bad feet,bad leg
After pasting the parameters, the user can further adjust other parameter settings at the bottom of the page, and you can also refer to the following figure to adjust. In addition, if you want to adjust the size of the generated picture, you can adjust the Width (width) and Height (height) here first.
After the parameters are set, click “Generate” to start generating images. Wait for a short while, and the picture will be complete! At this time, just long press to save the completed picture.
According to the parameter suggestions above, the resulting results refer to:
Sometimes the calculated pictures will not be directly displayed at the bottom of the page. At this time, we only need to go to the “Image Browser” and click the “First Page” button to see all the calculated pictures just now.
Then I will briefly share the test results of citing the LoRA extended model. The operation method is basically the same, just modify the parameters to regenerate the picture (parameter suggestions will also be attached below).
Prompt parameter reference (please paste the following parameters in the Prompt field):
Negative Prompt parameter reference (please paste the following parameters in the Negative Prompt field):
sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, bad anatomy,(long hair:1.4),DeepNegative,(fat:1.2),facing away, looking away,tilted head, {Multiple people}, lowres,bad anatomy,bad hands, text, error, missing fingers,extra digit, fewer digits, cropped, worstquality, low quality, normal quality,jpegartifacts,signature, watermark, username,blurry,bad feet,cropped,poorly drawn hands,poorly drawn face,mutation,deformed,worst quality,low quality,normal quality,jpeg artifacts,signature,watermark,extra fingers,fewer digits,extra limbs,extra arms,extra legs,malformed limbs,fused fingers,too many fingers,long neck,cross-eyed,mutated hands,polar lowres,bad body,bad proportions,gross proportions,text,error,missing fingers,missing arms,missing legs,extra digit, extra arms, extra leg, extra foot,
The results calculated with the above parameters are as follows:
*Reminder:
Since the Google Colab we use is free, the pictures will only be saved for a few hours, and even the host URL will expire after a period of non-use, so if you want to save it, you have to act fast (you can also pay for a subscription to support Google).
Most of these real-life models are based on real-life characters, so you can forget about it and look cool. But as long as it is “commercial” (such as: as a product promotion poster, open OnlyFans and other commercial activities), it is very likely to be sued. After all, brokerage companies in any country do not want their artist portraits to be misused, let alone the problem of confusing real ones, everyone should pay attention to it.