This is a captioning tool using ChatGPT (API) to describe images.
https://github.com/MNeMoNiCuZ/GPTCaption
Add your OpenAI API-key in the settings.ini-file.
Run main_gui.py.
Paste a bunch of URLs to images you wish to caption, one per line. Supports JPG, JPEG, PNG.
Optional: Edit the image analysis prompt.
Press Generate Captions
It will make a rough estimate of the API call costs and ask if you want to continue.
It will output everything in a date-folder, with an additional time-stamped folder each time you use the tool. Inside this folder there's also one caption with the same name as the folder that includes all captions.
You can use any service as long as the URL is public.
You can upload a batch of images (max 1000) on https://PostImages.org and get a list of URLs.
Be sure to select "Direct Link" as URL type.
Don't forget to edit the prompt to ChatGPT based on how you want to instruct it. It can give you the caption back in many different ways and styles.
1. 재게시된 모델의 권리는 원 제작자에게 있습니다.
2. 모델 원작자가 모델을 인증받으려면 공식 채널을 통해 SeaArt.AI 직원에게 문의하세요. 저희는 모든 창작자의 권리를 보호하기 위해 노력합니다. 인증하러 이동
