ComfyUI-Image-Evaluation

ComfyUI-Image-Evaluation
★ 9

图像评估相似度评分CLIP模型DINO模型
基于多模型对图像进行自动评估，支持CLIP文本评分、CLIP图像评分和DINO评分，便于量化相似度与质量对比。
💡 在ComfyUI中量化比较生成图像与文本或目标图像的相似度。
🍴 2 Forks💻 Python🔄 2024-12-06
📦
网盘下载
复制链接后前往夸克网盘下载
https://pan.quark.cn/s/a9fb3a59e10c
📦 requirements.txt
torchmetrics
Alt text
📄 README

ComfyUI Image Evaluation Node

This repository contains an extension to ComfyUI

This node allows you to evaluate the Clip Score between two images or a image and a text prompt using the CLIP model and DINO Score between two images using the DINO model.

Installation

  • Clone this repository into the custom_nodes folder of ComfyUI.
    • cd ComfyUI-Image-Evaluation
  • Install the requirements.
    • pip install -r requirements.txt
  • Restart ComfyUI and the extension should be loaded.

    • Features

  • Clip Text Score: Evaluate the Clip Score between two images or a image and a text prompt using the CLIP model.
  • Clip Image Score: Evaluate the Clip Score between a image and a target image using the CLIP model.
  • Dino Score: Evaluate the DINO Score between two images using the DINO model.

    • Nodes

    Author

  • Yujia Wu
  • GitHub: wu12023