Sign In
home
models
images
videos
posts
articles
bounties
tools
challenges
events
shop

Stable Diffusion : Grid-Based Analysis with Images and Short Videos (with Animatediff)

MisterMR's Avatar
MisterMR
Gold Writer Badge
Jan 19, 2024
generation guide
videos
Stable Diffusion : Grid-Based Analysis with Images and Short Videos (with Animatediff)

Objective

Generating different video with same prompt and different colors

Environment

For this test I will use:

Making video with this girl

Checkpoint

I will test to generate with 5 different checkpoints

Prompt

Positive

((Best Quality)), ((masterpiece)), DV_BUNNY_VICIOUS (detailed: 1.4), 3D, an image of a beautiful woman with cyberpunk blue eyes, black hair, vee dress, women wearing a corset mini dress, mini dress with pleated skirt, strapless dress, (black dress:1.6), HDR (High Dynamic Range),Ray Tracing,NVIDIA RTX,Super-Resolution,Unreal 5,Subsurface Dispersion, PBR Texture, Post-processing, Anisotropic filtering, Depth of field, Maximum clarity and sharpness, Multilayer textures, Albedo and specular maps, Surface shading, Accurate simulation of light-material interaction, Perfect Proportions, Octane Render, Two-Tone Lighting,Wide Aperture,Low ISO,White Balance,Rule of Thirds,8K RAW,

Negative

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

More details

Steps: 30, Sampler: Euler, CFG scale: 6, Seed: 284361537, Face restoration: GFPGAN, Size: 512x768, Model hash: 4d135d0197, Model: animedavincimix_v10, Clip skip: 2, Token merging ratio: 0.1, RNG: CPU, ADetailer model: face_yolov8n.pt, ADetailer prompt: DV_BUNNY_VICIOUS , 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: 24.1.1, TI hashes: "DV_Bunny_Vicious: 18dfe5177484", Noise multiplier: 0, Pad conds: True, Version: v1.7.0

Result

Some checkpoint had some issues to generate good images in red and black color.

Video Generation

Now I will enable the AnimateDiff option to see the results:

I leave the "number of frames to 0" it will generate 16 images for 2 seconds of video.

Converting mp4 to webp (if you forget in the settings)

I set the output in webp to import directly on this article (mp4 to webp covertion with FFMEG it is not the best).

If somebody wants a tip to convert mp4 to webp .. this works :

ffmpeg -i input_filename.mp4 -vcodec libwebp -filter:v fps=fps=20 -lossless 1 -loop 0 -preset default -an -vsync 0 -s 512:768 output_filename.webp

It takes in input input_filename.mp4 as input and save an output_filename.webp file.

(ffmpeg is a free tool you can download from here : https://ffmpeg.org/)

Video Results

Video AnimeDaVinciMix

Video Aniverse

Video epiCRealism

Video Matrix-Hentai-Plus

Video The Truality Engine PRO

As you can see with the same seed and different prompt animatediff works in a different for any checkpoint.

For a full mp4 video check here : https://civitai.com/posts/1234651

I hope you enjoy it. Comment if you have any question.

10

Comments