📸 GetImg

📸 GetImg: 一个 CLI 和 SDK，用于与 GetImg API 交互，通过各种端点实现图像生成和操作。

📖 目录

• 安装
• 功能
• 用法
• 选项
• 子命令
• 示例
• 贡献
• 许可证

🚀 安装

要安装 getimg CLI，请使用以下 Cargo 命令

cargo install --locked getimg --all-features

✨ 功能

• 从命令行界面与 GetImg API 交互。
• 使用各种端点生成和操作图像。
• 执行文本到图像和图像到图像的生成。
• 根据提示重新绘制图像或生成编辑后的图像。
• 利用 ControlNet 条件进行图像生成。

用法

在使用 getimg CLI 之前，请确保设置以下环境变量

export GETIMG_API_KEY=<your_getimg_api_key>

从 GetImg 控制台 生成 API 密钥。

⌨ 作为 CLI 使用

生成编辑后的图像

getimg edit -p "A man riding a horse on Mars." -i image.jpg -s 25 -g 7.5 -e 25 -y 1.5 -o png -n "Disfigured, cartoon, blurry" -c ddim

重新绘制图像

getimg paint -p "An image of a cityscape with neon lights." -i image.png -m edited_image.png -w 512 -a 512 -e 50 -s 5 -g 10.0 -o jpeg -c euler -f 1 -n "Disfigured, cartoon, blurry"

从文本生成图像

getimg t2i -p "A colorful sunset over the ocean." -w 512 -a 512 -s 5 -e 42 -o png -n "Disfigured, cartoon, blurry"

从另一图像生成图像

getimg i2i -p "Add a forest in the background." -i generated_image.png -s 6 -e 512 -o jpeg -f 0.5 -n "Disfigured, cartoon, blurry"

使用 ControlNet 条件生成图像

getimg cnet -p "A painting of a landscape." -i generated_image.png -f 1.0 -w 512 -a 512 -s 25 -g 7.5 -e 512 -c lms -o png -r canny-1.1 -n "Disfigured, cartoon, blurry"

🎨 选项

🛠 子命令

✨ 作为 Crate 使用

1. 添加 getimg Crate

   [dependencies]
   getimg = "0.0.1"
   

2. 使用提供的结构和方法在与 Rust 项目交互中使用 GetImg API。

   use anyhow::Result;
   use getimg::client::Client;
   use getimg::utils::save_image;
   use std::env;
   
   #[tokio::main]
   async fn main() -> Result<()> {
       // Example usage
       let api_key = env::var("GETIMG_API_KEY").unwrap_or_default().to_owned();
       let text_prompt = "a kanban-style task board with three columns: To Do, In Progress, and Done. Each column contains a list of tasks, with the To Do column having the most tasks and the Done column having the fewest. The tasks are color-coded, with the To Do tasks being red, the In Progress tasks being yellow, and the Done tasks being green. The board also has a header that includes the title \"To Do List\" and a search bar.";
       let negative_prompt = Some("Disfigured, cartoon, blurry");
   
       // Create a new instance of the GetIMG Client
       let mut client = Client::new(&api_key, "lcm-realistic-vision-v5-1");
       println!("{:?}", client);
   
       // Generate image from text prompt
       let text_response = client
           .generate_image_from_text(
               text_prompt,
               1024,
               1024,
               4,
               "jpeg",
               negative_prompt,
               Some(512),
           )
           .await?;
   
       // Save text response image to file
       save_image(&text_response.image, "text_response.jpg")?;
   
       // Generate image from image prompt
       let image_response = client
           .generate_image_from_image(
               "a photo of an astronaut riding a horse on mars",
               &text_response.image,
               4,
               512,
               "jpeg",
               negative_prompt,
               Some(0.5),
           )
           .await?;
   
       // Save image response image to file
       save_image(&image_response.image, "image_response.png")?;
   
       let controlnet_response = client
           .generate_image_using_controlnet(
               "softedge-1.1",
               "a photo of an astronaut riding a horse on mars",
               "Disfigured, cartoon, blurry",
               &image_response.image,
               1.0,
               512,
               512,
               25,
               7.5,
               512,
               "euler",
               "png",
           )
           .await?;
   
       save_image(&controlnet_response.image, "controlnet_response.png")?;
   
       // Generate repainted image
       let repainted_image_response = client
           .generate_repainted_image(
               text_prompt,
               negative_prompt,
               &text_response.image,
               &image_response.image,
               Some(1.0),
               512,
               512,
               25,
               7.5,
               512,
               "euler",
               "png",
           )
           .await?;
   
       // Save repainted image to file
       save_image(&repainted_image_response.image, "repainted_image.png")?;
   
       // Generate edited image
       let edited_image_response = client
           .generate_edited_image(
               text_prompt,
               negative_prompt,
               &repainted_image_response.image,
               1.5,
               25,
               7.5,
               54,
               "ddim",
               "png",
           )
           .await?;
   
       // Save edited image to file
       save_image(&edited_image_response.image, "edited_image.png")?;
   
       Ok(())
   }
   

📄 许可证

本项目根据 MIT 许可证 许可。