Skip to content

Couple of stable diffusion tools I use with command line alias's

Notifications You must be signed in to change notification settings

WSINTRA/sd-tools

Repository files navigation

Declare an empty array

declare -a array

Read the contents of the text file line by line

while read line; do
  arr+=("$line")
done < output.txt
Once the array has been created, you can use it to feed the scene descriptions into your custom AI image generator, one description at a time. You can do this using a loop, as shown below:
# Loop through each element in the array
for description in "${array[@]}"; do
  # Feed the description into the AI image generator

  ./stable_diffusion_ai "$description"
done
# Iterate over the array using indices
for i in $(seq 1 $((${#arr[@]} - 1))); do
  prompt_1=${arr[$i]}
  prompt_2=${arr[$(($i + 1))]}
  python interpolation_steps.py --prompt_1="$prompt_1" --prompt_2="$prompt_2" --index=$i --filename="ChatGPT"
done

for element in "${arr[@]}"; do echo "$element" done

You can use ffmpeg to create an MP4 video from a folder of GIF files. Here is an example command that you can use:

for f in *.gif ; do echo file '$f' >> fileList.txt;

ffmpeg -f concat -safe 0 -i fileList.txt -c copy mergedVideo.mp4 ffmpeg -f concat -safe 0 -i fileList.txt -c:v libx264 -preset ultrafast -qp 0 output.mkv ffmpeg -i output.mkv -vcodec libx264 -pix_fmt yuv420p movie.mp4 ffmpeg -f concat -safe 0 -i fileList.txt -vcodec libx264 -pix_fmt yuv420p movie.mp4

Notes to self. Large interpolation steps did not produce better results. Still need to work on the zooming through scenes.

declare -a arr

while read line; do arr+=("$line") done < noSpace.txt

for element in "${arr[@]}"; do echo "$element" done

for i in $(seq 1 $((${#arr[@]} - 1))); do prompt_1=${arr[$i]} prompt_2=${arr[$(($i + 1))]} python interpolation_steps.py --prompt_1="$prompt_1" --prompt_2="$prompt_2" --index=$i --filename="AZ" done

openai api completions.create -m text-davinci-003 -p "therefore we follow only nature" -t 0 -M 700 --stream >> file.txt

Other cool ideas

use ffmpeg to extract frames from a short clip then use AI to alter each frame, maybe image to image, turn it into cartoon or something, then use ffmpeg to re encode the frames

About

Couple of stable diffusion tools I use with command line alias's

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages